path: root/lib/ssh/doc/src/ssh_sftp.xml

<?xml version="1.0" encoding="utf-8" ?>
<!DOCTYPE erlref SYSTEM "erlref.dtd">

<erlref>
  <header>
    <copyright>
      <year>2005</year><year>2018</year>
      <holder>Ericsson AB. All Rights Reserved.</holder>
    </copyright>
    <legalnotice>
      Licensed under the Apache License, Version 2.0 (the "License");
      you may not use this file except in compliance with the License.
      You may obtain a copy of the License at
 
          http://www.apache.org/licenses/LICENSE-2.0

      Unless required by applicable law or agreed to in writing, software
      distributed under the License is distributed on an "AS IS" BASIS,
      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
      See the License for the specific language governing permissions and
      limitations under the License.

    </legalnotice>

    <title>ssh_sftp</title>
    <prepared>OTP</prepared>
    <docno></docno>
    <date>2005-09-22</date>
    <rev></rev>
    <file>ssh_sftp.sgml</file>
  </header>
  <module since="">ssh_sftp</module>
  <modulesummary>SFTP client.</modulesummary>
  <description>
    <p>This module implements an SSH FTP (SFTP) client. SFTP is a
    secure, encrypted file transfer service available for
    SSH.</p>
  </description>
  
  <datatypes>
    <datatype_title>Error cause</datatype_title>
    <datatype>
      <name name="reason"/>
      <desc>
	<p>A description of the reason why an operation failed.</p>
	<p>The <c>atom()</c> value is formed from the sftp error codes in the protocol-level responses as defined in 
	<url href="https://tools.ietf.org/html/draft-ietf-secsh-filexfer-13#page-49">draft-ietf-secsh-filexfer-13</url>
	section 9.1.
	  The codes are named as <c>SSH_FX_*</c> which are transformed into lowercase of the star-part.
	  E.g. the error code <c>SSH_FX_NO_SUCH_FILE</c>
	  will cause the <c>reason()</c> to be <c>no_such_file</c>.
	</p>
	<p>The <c>string()</c> reason is the error information from the server in case of an exit-signal.  If that information is empty, the reason is the exit signal name.
	</p>
	<p>The <c>tuple()</c> reason are other errors like for example <c>{exit_status,1}</c>.
	</p>
      </desc>
    </datatype>

    <datatype_title>Crypto operations for open_tar</datatype_title>
    <datatype>
      <name name="tar_crypto_spec"/>
      <name name="encrypt_spec"/>
      <name name="decrypt_spec"/>
      <desc>
	<p>Specifies the encryption or decryption applied to tar files when using
	<seealso marker="#open_tar/3">open_tar/3</seealso> or 
	<seealso marker="#open_tar/4">open_tar/4</seealso>.
	</p>
	<p>The encryption or decryption is applied to the generated stream of
	bytes prior to sending the resulting stream to the SFTP server.
	</p>
	<p>For code examples see Section
	<seealso marker="using_ssh#example-with-encryption">Example with encryption</seealso>
	in the ssh Users Guide.
	</p>
      </desc>
    </datatype>

    <datatype>
      <name name="init_fun"/>
      <name name="chunk_size"/>
      <name name="crypto_state"/>
      <desc>
	<p>The <c>init_fun()</c> in the
	<seealso marker="#type-tar_crypto_spec">tar_crypto_spec</seealso>
	is applied once prior to any other <c>crypto</c>
	operation. The intention is that this function initiates the encryption or
	decryption for example by calling
	<seealso marker="crypto:crypto#crypto_init/4">crypto:crypto_init/4</seealso>
	or similar. The <c>crypto_state()</c> is the state such a function may return.
	</p>
	<p>If the selected cipher needs to have the input data partioned into
	blocks of a certain size, the <c>init_fun()</c> should return the second
	form of return value with the <c>chunk_size()</c> set to the block size.
	If the <c>chunk_size()</c> is <c>undefined</c>, the size of the <c>PlainBin</c>s varies,
	because this is	intended for stream crypto, whereas a fixed <c>chunk_size()</c> is intended for block crypto.
	A <c>chunk_size()</c> can be changed in the return from the <c>crypto_fun()</c>.
	The value can be changed between <c>pos_integer()</c> and <c>undefined</c>.
	</p>
      </desc>
    </datatype>

    <datatype>
      <name name="crypto_fun"/>
      <name name="crypto_result"/>
      <desc>
	<p>The initial <c>crypto_state()</c> returned from the
	<seealso marker="#type-init_fun">init_fun()</seealso>
	is folded into repeated applications of the <c>crypto_fun()</c> in the
	<seealso marker="#type-tar_crypto_spec">tar_crypto_spec</seealso>.
	The binary returned from that fun is sent to the remote SFTP server and
	the new <c>crypto_state()</c> is used in the next call of the
	<c>crypto_fun()</c>.
	</p>
	<p>If the <c>crypto_fun()</c> reurns a <c>chunk_size()</c>, that value
	is as block size for further blocks in calls to <c>crypto_fun()</c>.
	</p>
      </desc>
    </datatype>

    <datatype>
      <name name="final_fun"/>
      <desc>
	<p>If doing encryption,
	the <c>final_fun()</c>  in the
	<seealso marker="#type-tar_crypto_spec">tar_crypto_spec</seealso>
	is applied to the last piece of data.
	The <c>final_fun()</c> is responsible for padding (if needed) and
	encryption of that last piece.
	</p>
      </desc>
    </datatype>
  </datatypes>

  <funcs>
     <func>
       <name name="apread" arity="4" since=""/>
       <fsummary>Reads asynchronously from an open file.</fsummary>
       <desc><p>The <c><![CDATA[apread/4]]></c> function reads from a specified position,
       combining the <seealso marker="#position-3"><c>position/3</c></seealso> and 
       <seealso marker="#aread-3"><c>aread/3</c></seealso> functions.</p>
       </desc>
     </func>
     
     <func>
	<name name="apwrite" arity="4" since=""/>
	<fsummary>Writes asynchronously to an open file.</fsummary>
       <desc><p>The <c><![CDATA[apwrite/4]]></c> function writes to a specified position,
       combining the <seealso marker="#position-3"><c>position/3</c></seealso> and 
       <seealso marker="#awrite-3"><c>awrite/3</c></seealso> functions.</p>
       </desc>
      </func>
      
      <func>
	<name name="aread" arity="3" since=""/>
	<fsummary>Reads asynchronously from an open file.</fsummary>
	<desc>
	  <p>Reads from an open file, without waiting for the result. If the
          handle is valid, the function returns <c><![CDATA[{async, N}]]></c>, where <c>N</c>
          is a term guaranteed to be unique between calls of <c><![CDATA[aread]]></c>.
          The actual data is sent as a message to the calling process. This
          message has the form <c><![CDATA[{async_reply, N, Result}]]></c>, where
          <c><![CDATA[Result]]></c> is the result from the read, either <c><![CDATA[{ok, Data}]]></c>,
          <c><![CDATA[eof]]></c>, or <c><![CDATA[{error, reason()}]]></c>.</p>
      </desc>
    </func>

    <func>
      <name name="awrite" arity="3" since=""/>
      <fsummary>Writes asynchronously to an open file.</fsummary>
      <desc>
        <p>Writes to an open file, without waiting for the result. If the
          handle is valid, the function returns <c><![CDATA[{async, N}]]></c>, where <c>N</c>
          is a term guaranteed to be unique between calls of
          <c><![CDATA[awrite]]></c>. The result of the <c><![CDATA[write]]></c> operation is sent
          as a message to the calling process. This message has the form
          <c><![CDATA[{async_reply, N, Result}]]></c>, where <c><![CDATA[Result]]></c> is the result
          from the write, either <c><![CDATA[ok]]></c>, or <c><![CDATA[{error, reason()}]]></c>.</p>
      </desc>
    </func>

    <func>
      <name name="close" arity="2" since=""/>
      <name name="close" arity="3" since=""/>
      <fsummary>Closes an open handle.</fsummary>
      <desc>
        <p>Closes a handle to an open file or directory on the server.</p>
      </desc>
    </func>

    <func>
      <name name="delete" arity="2" since=""/>
      <name name="delete" arity="3" since=""/>
      <fsummary>Deletes a file.</fsummary>
      <desc>
        <p>Deletes the file specified by <c><![CDATA[Name]]></c>.
	</p>
      </desc>
    </func>

    <func>
      <name name="del_dir" arity="2" since=""/>
      <name name="del_dir" arity="3" since=""/>
      <fsummary>Deletes an empty directory.</fsummary>
      <desc>
	<p>Deletes a directory specified by <c><![CDATA[Name]]></c>.
	The directory must be empty before it can be successfully deleted.
	</p>
      </desc>
    </func>

     <func>
      <name name="list_dir" arity="2" since=""/>
      <name name="list_dir" arity="3" since=""/>
      <fsummary>Lists the directory.</fsummary>
      <desc>
        <p>Lists the given directory on the server, returning the
          filenames as a list of strings.</p>
      </desc>
    </func>

    <func>
      <name name="make_dir" arity="2" since=""/>
      <name name="make_dir" arity="3" since=""/>
      <fsummary>Creates a directory.</fsummary>
      <desc>
        <p>Creates a directory specified by <c><![CDATA[Name]]></c>. <c><![CDATA[Name]]></c>
          must be a full path to a new directory. The directory can only be
	created in an existing directory.</p>
      </desc>
    </func>

    <func>
      <name name="make_symlink" arity="3" since=""/>
      <name name="make_symlink" arity="4" since=""/>
      <fsummary>Creates a symbolic link.</fsummary>
      <desc>
        <p>Creates a symbolic link pointing to <c><![CDATA[Target]]></c> with the
          name <c><![CDATA[Name]]></c>.
	</p>
      </desc>
    </func>

    <func>
      <name name="open" arity="3" since=""/>
      <name name="open" arity="4" since=""/>
      <fsummary>Opens a file and returns a handle.</fsummary>
      <desc>
        <p>Opens a file on the server and returns a handle, which
          can be used for reading or writing.</p>
      </desc>
    </func>
    <func>
      <name name="opendir" arity="2" since=""/>
      <name name="opendir" arity="3" since=""/>
      <fsummary>Opens a directory and returns a handle.</fsummary>
      <desc>
        <p>Opens a handle to a directory on the server. The handle
          can be used for reading directory contents.</p>
      </desc>
    </func>

    <func>
      <name name="open_tar" arity="3" since="OTP 17.4"/>
      <name name="open_tar" arity="4" since="OTP 17.4"/>
      <fsummary>Opens a tar file on the server to which <c>ChannelPid</c>
      is connected and returns a handle.</fsummary>
      <desc>
        <p>Opens a handle to a tar file on the server, associated with <c>ChannelPid</c>.
	The handle can be used for remote tar creation and extraction. The actual writing
	and reading is performed by calls to 
	<seealso marker="stdlib:erl_tar#add-3">erl_tar:add/3,4</seealso> and
	<seealso marker="stdlib:erl_tar#extract-2">erl_tar:extract/2</seealso>.
	Note: The 
	<seealso marker="stdlib:erl_tar#init-3">erl_tar:init/3</seealso> function should not
	be called, that one is called by this open_tar function.
	</p>
	<p>For code examples see Section
	<seealso marker="using_ssh#sftp-client-with-tar-compression">SFTP Client with TAR Compression</seealso>
	in the ssh Users Guide.
	</p>
	<p>The <c>crypto</c> mode option is explained in the data types section above, see
	<seealso marker="#Crypto operations for open_tar">Crypto operations for open_tar</seealso>.
	Encryption is assumed if the <c>Mode</c> contains <c>write</c>, and
	decryption if the <c>Mode</c> contains <c>read</c>.
	</p>
      </desc>
    </func>

    <func>
      <name name="position" arity="3" since=""/>
      <name name="position" arity="4" since=""/>
      <fsummary>Sets the file position of a file.</fsummary>
      <desc>
        <p>Sets the file position of the file referenced by <c><![CDATA[Handle]]></c>.
          Returns <c><![CDATA[{ok, NewPosition}]]></c> (as an absolute offset) if
          successful, otherwise <c><![CDATA[{error, reason()}]]></c>. <c><![CDATA[Location]]></c> is
          one of the following:</p>
        <taglist>
          <tag><c><![CDATA[Offset]]></c></tag>
          <item>
            <p>The same as <c><![CDATA[{bof, Offset}]]></c>.</p>
          </item>
          <tag><c><![CDATA[{bof, Offset}]]></c></tag>
          <item>
            <p>Absolute offset.</p>
          </item>
          <tag><c><![CDATA[{cur, Offset}]]></c></tag>
          <item>
            <p>Offset from the current position.</p>
          </item>
          <tag><c><![CDATA[{eof, Offset}]]></c></tag>
          <item>
            <p>Offset from the end of file.</p>
          </item>
          <tag><c><![CDATA[bof | cur | eof]]></c></tag>
          <item>
            <p>The same as eariler with <c><![CDATA[Offset]]></c> 0,
	    that is, <c><![CDATA[{bof, 0} | {cur, 0} | {eof, 0}]]></c>.
	    </p>
          </item>
        </taglist>
      </desc>
    </func>

    <func>
      <name name="pread" arity="4" since=""/>
      <name name="pread" arity="5" since=""/>
      <fsummary>Reads from an open file.</fsummary>
       <desc><p>The <c><![CDATA[pread/3,4]]></c> function reads from a specified position,
       combining the <seealso marker="#position-3"><c>position/3</c></seealso> and 
       <seealso marker="#read-3"><c>read/3,4</c></seealso> functions.</p>
       </desc>
     </func>

    <func>
      <name name="pwrite" arity="4" since=""/>
      <name name="pwrite" arity="5" since=""/>
      <fsummary>Writes to an open file.</fsummary>
       <desc><p>The <c><![CDATA[pwrite/3,4]]></c> function writes to a specified position,
       combining the <seealso marker="#position-3"><c>position/3</c></seealso> and 
       <seealso marker="#write-3"><c>write/3,4</c></seealso> functions.</p>
       </desc>
    </func>

    <func>
      <name name="read" arity="3" since=""/>
      <name name="read" arity="4" since=""/>
      <fsummary>Reads from an open file.</fsummary>
      <desc>
        <p>Reads <c><![CDATA[Len]]></c> bytes from the file referenced by
          <c><![CDATA[Handle]]></c>. Returns <c><![CDATA[{ok, Data}]]></c>, <c><![CDATA[eof]]></c>, or
          <c><![CDATA[{error, reason()}]]></c>. If the file is opened with <c><![CDATA[binary]]></c>,
          <c><![CDATA[Data]]></c> is a binary, otherwise it is a string.</p>
        <p>If the file is read past <c>eof</c>, only the remaining bytes
          are read and returned. If no bytes are read, <c><![CDATA[eof]]></c>
          is returned.</p>
      </desc>
    </func>

    <func>
      <name name="read_file" arity="2" since=""/>
      <name name="read_file" arity="3" since=""/>
      <fsummary>Reads a file.</fsummary>
      <desc>
        <p>Reads a file from the server, and returns the data in a binary.</p>
      </desc>
    </func>

    <func>
      <name name="read_file_info" arity="2" since=""/>
      <name name="read_file_info" arity="3" since=""/>
      <fsummary>Gets information about a file.</fsummary>
      <desc>
        <p>Returns a <c><![CDATA[file_info]]></c> record from the file system object specified by
          <c><![CDATA[Name]]></c> or <c><![CDATA[Handle]]></c>. See
	<seealso marker="kernel:file#read_file_info-2">file:read_file_info/2</seealso>
	for information about the record.
	</p>
	<p>
	  Depending on the underlying OS:es links might be followed and info on the final file, directory
	  etc is returned. See <seealso marker="#read_link_info-2">read_link_info/2</seealso>
	  on how to get information on links instead.
	</p>
      </desc>
    </func>

     <func>
      <name name="read_link" arity="2" since=""/>
      <name name="read_link" arity="3" since=""/>
      <fsummary>Reads symbolic link.</fsummary>
      <desc>
        <p>Reads the link target from the symbolic link specified by <c><![CDATA[name]]></c>.
	</p>
      </desc>
    </func>

    <func>
      <name since="" name="read_link_info" arity="2"/>
      <name since="" name="read_link_info" arity="3"/>
      <fsummary>Gets information about a symbolic link.</fsummary>
      <desc>
        <p>Returns a <c><![CDATA[file_info]]></c> record from the symbolic
          link specified by <c><![CDATA[Name]]></c> or <c><![CDATA[Handle]]></c>.
	  See
	  <seealso marker="kernel:file#read_link_info-2">file:read_link_info/2</seealso>
	  for information about the record.
	</p>
      </desc>
    </func>

    <func>
      <name since="" name="rename" arity="3"/>
      <name since="" name="rename" arity="4"/>
      <fsummary>Renames a file.</fsummary>
      <desc>
        <p>Renames a file named <c><![CDATA[OldName]]></c> and gives it the name
          <c><![CDATA[NewName]]></c>.
	</p>
      </desc>
    </func>

    <func>
      <name since="">start_channel(ConnectionRef) -></name>
      <name since="">start_channel(ConnectionRef, Options) ->
      {ok, Pid} | {error, reason()|term()}</name>

      <name since="">start_channel(Host, Options) -></name>
      <name since="">start_channel(Host, Port, Options) -> 
      {ok, Pid, ConnectionRef} | {error, reason()|term()}</name>

      <name since="">start_channel(TcpSocket) -></name>
      <name since="">start_channel(TcpSocket, Options) ->
      {ok, Pid, ConnectionRef} | {error, reason()|term()}</name>

      <fsummary>Starts an SFTP client.</fsummary>
      <type>
        <v>Host = string()</v>
        <v>ConnectionRef = connection_ref()</v>
        <v>Port = integer()</v>
	<v>TcpSocket = port()</v>
	<d>The socket is supposed to be from <seealso marker="kernel:gen_tcp#connect-3">gen_tcp:connect</seealso> or <seealso marker="kernel:gen_tcp#accept-1">gen_tcp:accept</seealso> with option <c>{active,false}</c></d>
        <v>Options = [{Option, Value}]</v>
      </type>
      <desc>
	<p>If no connection reference is provided, a connection is set
	up, and the new connection is returned. An SSH channel process
	is started to handle the communication with the SFTP server.
	The returned <c>pid</c> for this process is to be used as input to
	all other API functions in this module.</p>

	<p>Options:</p>
	<taglist>
          <tag><c><![CDATA[{timeout, timeout()}]]></c></tag>
	  <item>
	    <p>There are two ways to set a timeout for the underlying ssh connection:</p>
	    <list>
	      <item>If the connection timeout option <c>connect_timeout</c> is set, that value
	      is used also for the negotiation timeout and this option (<c>timeout</c>) is ignored.</item>
	      <item>Otherwise, this option (<c>timeout</c>) is used as the negotiation timeout
	      only and there is no connection timeout set</item>
	    </list>
	    <p>The value defaults to <c>infinity</c>.</p>
	  </item>
	  <tag>
	    <c><![CDATA[{sftp_vsn, integer()}]]></c>
	  </tag>
	  <item>
	    <p>
	      Desired SFTP protocol version.
	      The actual version is the minimum of
	      the desired version and the maximum supported
	      versions by the SFTP server.
	    </p>
	  </item>
	</taglist>
	<p>All other options are directly passed to
	<seealso marker="ssh">ssh:connect/3</seealso> or ignored if a
	connection is already provided.</p>
      </desc>
    </func>

     <func>
      <name since="" name="stop_channel" arity="1"/>
      <fsummary>Stops the SFTP client channel.</fsummary>
      <desc>
	<p>Stops an SFTP channel. Does not close the SSH connection.
	Use <seealso marker="ssh#close-1">ssh:close/1</seealso> to close it.</p>
      </desc>
    </func>

    <func>
      <name since="" name="write" arity="3"/>
      <name since="" name="write" arity="4"/>
      <fsummary>Writes to an open file.</fsummary>
      <desc>
        <p>Writes <c><![CDATA[data]]></c> to the file referenced by <c><![CDATA[Handle]]></c>.
	The file is to be opened with <c><![CDATA[write]]></c> or <c><![CDATA[append]]></c>
	flag. Returns <c><![CDATA[ok]]></c> if successful or <c><![CDATA[{error, reason()}]]></c>
	otherwise.</p>
      </desc>
    </func>
    
    <func>
      <name since="" name="write_file" arity="3"/>
      <name since="" name="write_file" arity="4"/>
      <fsummary>Writes a file.</fsummary>
      <desc>
        <p>Writes a file to the server.	The file is created if it does not exist
	but overwritten if it exists.</p>
      </desc>
    </func>
    
    <func>
      <name since="" name="write_file_info" arity="3"/>
      <name since="" name="write_file_info" arity="4"/>
      <fsummary>Writes information for a file.</fsummary>
      <desc>
        <p>Writes file information from a <c><![CDATA[file_info]]></c> record to the
	file specified by <c><![CDATA[Name]]></c>. See
	<seealso marker="kernel:file#write_file_info-2">file:write_file_info/[2,3]</seealso>
	for information about the record.
	</p>
      </desc>
    </func>
  </funcs>
  
</erlref>