.\" Automatically generated by Podwrapper::Man 1.24.1 (Pod::Simple 3.40)
.\"
.\" Standard preamble:
.\" ========================================================================
.de Sp \" Vertical space (when we can't use .PP)
.if t .sp .5v
.if n .sp
..
.de Vb \" Begin verbatim text
.ft CW
.nf
.ne \\$1
..
.de Ve \" End verbatim text
.ft R
.fi
..
.\" Set up some character translations and predefined strings.  \*(-- will
.\" give an unbreakable dash, \*(PI will give pi, \*(L" will give a left
.\" double quote, and \*(R" will give a right double quote.  \*(C+ will
.\" give a nicer C++.  Capital omega is used to do unbreakable dashes and
.\" therefore won't be available.  \*(C` and \*(C' expand to `' in nroff,
.\" nothing in troff, for use with C<>.
.tr \(*W-
.ds C+ C\v'-.1v'\h'-1p'\s-2+\h'-1p'+\s0\v'.1v'\h'-1p'
.ie n \{\
.    ds -- \(*W-
.    ds PI pi
.    if (\n(.H=4u)&(1m=24u) .ds -- \(*W\h'-12u'\(*W\h'-12u'-\" diablo 10 pitch
.    if (\n(.H=4u)&(1m=20u) .ds -- \(*W\h'-12u'\(*W\h'-8u'-\"  diablo 12 pitch
.    ds L" ""
.    ds R" ""
.    ds C` ""
.    ds C' ""
'br\}
.el\{\
.    ds -- \|\(em\|
.    ds PI \(*p
.    ds L" ``
.    ds R" ''
.    ds C`
.    ds C'
'br\}
.\"
.\" Escape single quotes in literal strings from groff's Unicode transform.
.ie \n(.g .ds Aq \(aq
.el       .ds Aq '
.\"
.\" If the F register is >0, we'll generate index entries on stderr for
.\" titles (.TH), headers (.SH), subsections (.SS), items (.Ip), and index
.\" entries marked with X<> in POD.  Of course, you'll have to process the
.\" output yourself in some meaningful fashion.
.\"
.\" Avoid warning from groff about undefined register 'F'.
.de IX
..
.nr rF 0
.if \n(.g .if rF .nr rF 1
.if (\n(rF:(\n(.g==0)) \{\
.    if \nF \{\
.        de IX
.        tm Index:\\$1\t\\n%\t"\\$2"
..
.        if !\nF==2 \{\
.            nr % 0
.            nr F 2
.        \}
.    \}
.\}
.rr rF
.\" ========================================================================
.\"
.IX Title "nbdkit-python-plugin 3"
.TH nbdkit-python-plugin 3 "2021-01-20" "nbdkit-1.24.1" "NBDKIT"
.\" For nroff, turn off justification.  Always turn off hyphenation; it makes
.\" way too many mistakes in technical documents.
.if n .ad l
.nh
.SH "NAME"
nbdkit\-python\-plugin \- nbdkit python plugin
.SH "SYNOPSIS"
.IX Header "SYNOPSIS"
.Vb 1
\& nbdkit python /path/to/plugin.py [arguments...]
.Ve
.SH "DESCRIPTION"
.IX Header "DESCRIPTION"
\&\f(CW\*(C`nbdkit\-python\-plugin\*(C'\fR is an embedded Python interpreter for
\&\fBnbdkit\fR\|(1), allowing you to write nbdkit plugins in Python.
.SS "If you have been given an nbdkit Python plugin"
.IX Subsection "If you have been given an nbdkit Python plugin"
Assuming you have a Python script which is an nbdkit plugin, you run it
like this:
.PP
.Vb 1
\& nbdkit python /path/to/plugin.py
.Ve
.PP
You may have to add further \f(CW\*(C`key=value\*(C'\fR arguments to the command
line.  Read the Python script to see if it requires any.
.SH "WRITING A PYTHON NBDKIT PLUGIN"
.IX Header "WRITING A PYTHON NBDKIT PLUGIN"
For example plugins written in Python, see:
https://github.com/libguestfs/nbdkit/blob/master/plugins/python/examples
.PP
Broadly speaking, Python nbdkit plugins work like C ones, so you should
read \fBnbdkit\-plugin\fR\|(3) first.
.PP
To write a Python nbdkit plugin, you create a Python file which
contains at least the following required functions (in the top level
\&\f(CW\*(C`_\|_main_\|_\*(C'\fR module):
.PP
.Vb 7
\& API_VERSION = 2
\& def open(readonly):
\&   # see below
\& def get_size(h):
\&   # see below
\& def pread(h, buf, offset, flags):
\&   # see below
.Ve
.PP
Note that the subroutines must have those literal names (like \f(CW\*(C`open\*(C'\fR),
because the C part looks up and calls those functions directly.  You
may want to include documentation and globals (eg. for storing global
state).  Any other top level statements are run when the script is
loaded, just like ordinary Python.
.SS "Python versions"
.IX Subsection "Python versions"
In \fBnbdkit ≤ 1.14\fR, either Python 2 or 3 could be used.  It was
selected at compile time by either:
.PP
.Vb 1
\& ./configure
.Ve
.PP
which selected the version of Python by looking at the \f(CW\*(C`python\*(C'\fR
interpreter found on the \f(CW$PATH\fR.  Or:
.PP
.Vb 1
\& ./configure PYTHON=/usr/bin/python3
.Ve
.PP
which allowed you to select a different interpreter and hence a
different version of Python.
.PP
\&\fBnbdkit ≥ 1.16\fR drops all support for Python 2, since Python 2
has reached its end of life.
.PP
The new behaviour is that \f(CW\*(C`./configure\*(C'\fR looks for \f(CW\*(C`python3\*(C'\fR or
\&\f(CW\*(C`python\*(C'\fR (in that order) on the \f(CW$PATH\fR.  It will fail if the first
interpreter it finds is a Python 2 interpreter.  You may also still
choose a Python interpreter by setting the \f(CW\*(C`PYTHON\*(C'\fR variable at
configure time as above.
.PP
If you wish to continue using nbdkit plugins written in Python 2 then
you must use nbdkit ≤ 1.14, but we would advise you to update your
plugins.
.PP
To find out which version the Python plugin was compiled for, use the
\&\fI\-\-dump\-plugin\fR option, eg:
.PP
.Vb 4
\& $ nbdkit python \-\-dump\-plugin
\& ...
\& python_version=3.7.0
\& python_pep_384_abi_version=3
.Ve
.SS "\s-1API\s0 versions"
.IX Subsection "API versions"
The nbdkit \s-1API\s0 has evolved and new versions are released periodically.
To ensure backwards compatibility plugins have to opt in to the new
version.  From Python you do this by declaring a constant in your
module:
.PP
.Vb 1
\& API_VERSION = 2
.Ve
.PP
(where 2 is the latest version at the time this documentation was
written).  All newly written Python modules must have this constant.
.SS "Executable script"
.IX Subsection "Executable script"
If you want you can make the script executable and include a \*(L"shebang\*(R"
at the top:
.PP
.Vb 1
\& #!/usr/sbin/nbdkit python
.Ve
.PP
See also \*(L"Shebang scripts\*(R" in \fBnbdkit\fR\|(1).
.PP
These scripts can also be installed in the \f(CW$plugindir\fR.  See
\&\*(L"\s-1WRITING PLUGINS IN OTHER PROGRAMMING LANGUAGES\*(R"\s0 in \fBnbdkit\-plugin\fR\|(3).
.SS "Methods"
.IX Subsection "Methods"
Your script may use \f(CW\*(C`import nbdkit\*(C'\fR to have access to the following
methods in the \f(CW\*(C`nbdkit\*(C'\fR module:
.PP
\fI\f(CI\*(C`nbdkit.debug(msg)\*(C'\fI\fR
.IX Subsection "nbdkit.debug(msg)"
.PP
Send a debug message to stderr or syslog if verbose messages are
enabled.
.PP
\fI\f(CI\*(C`nbdkit.export_name()\*(C'\fI\fR
.IX Subsection "nbdkit.export_name()"
.PP
Return the export name negotiated with the client as a Unicode string.
Note this should not be trusted because the client can send whatever
it wants.
.PP
\fI\f(CI\*(C`nbdkit.set_error(err)\*(C'\fI\fR
.IX Subsection "nbdkit.set_error(err)"
.PP
Record \f(CW\*(C`err\*(C'\fR as the reason you are about to throw an exception. \f(CW\*(C`err\*(C'\fR
should correspond to usual errno values, where it may help to
\&\f(CW\*(C`import errno\*(C'\fR.
.SS "Exceptions"
.IX Subsection "Exceptions"
Python callbacks should throw exceptions to indicate errors.  Remember
to use \f(CW\*(C`nbdkit.set_error\*(C'\fR if you need to control which error is sent
back to the client; if omitted, the client will see an error of \f(CW\*(C`EIO\*(C'\fR.
.SS "Python callbacks"
.IX Subsection "Python callbacks"
This just documents the arguments to the callbacks in Python, and any
way that they differ from the C callbacks.  In all other respects they
work the same way as the C callbacks, so you should go and read
\&\fBnbdkit\-plugin\fR\|(3).
.ie n .IP """dump_plugin""" 4
.el .IP "\f(CWdump_plugin\fR" 4
.IX Item "dump_plugin"
(Optional)
.Sp
There are no arguments or return value.
.ie n .IP """config""" 4
.el .IP "\f(CWconfig\fR" 4
.IX Item "config"
(Optional)
.Sp
.Vb 2
\& def config(key, value):
\&   # no return value
.Ve
.ie n .IP """config_complete""" 4
.el .IP "\f(CWconfig_complete\fR" 4
.IX Item "config_complete"
(Optional)
.Sp
There are no arguments or return value.
.ie n .IP """thread_model""" 4
.el .IP "\f(CWthread_model\fR" 4
.IX Item "thread_model"
(Optional, nbdkit ≥ 1.22)
.Sp
.Vb 2
\& def thread_model():
\&   return nbdkit.THEAD_MODEL_SERIALIZE_ALL_REQUESTS
.Ve
.Sp
See \*(L"Threads\*(R" below.
.ie n .IP """get_ready""" 4
.el .IP "\f(CWget_ready\fR" 4
.IX Item "get_ready"
(Optional)
.Sp
There are no arguments or return value.
.ie n .IP """list_exports""" 4
.el .IP "\f(CWlist_exports\fR" 4
.IX Item "list_exports"
(Optional)
.Sp
.Vb 4
\& def list_exports(readonly, is_tls):
\&   # return an iterable object (eg. list) of
\&   # (name, description) tuples or bare names:
\&   return [ (name1, desc1), name2, (name3, desc3), ... ]
.Ve
.ie n .IP """default_export""" 4
.el .IP "\f(CWdefault_export\fR" 4
.IX Item "default_export"
(Optional)
.Sp
.Vb 3
\& def default_export(readonly, is_tls):
\&   # return a string
\&   return "name"
.Ve
.ie n .IP """open""" 4
.el .IP "\f(CWopen\fR" 4
.IX Item "open"
(Required)
.Sp
.Vb 2
\& def open(readonly):
\&   # return handle
.Ve
.Sp
You can return any non-NULL Python value as the handle.  It is passed
back in subsequent calls.
.ie n .IP """close""" 4
.el .IP "\f(CWclose\fR" 4
.IX Item "close"
(Optional)
.Sp
.Vb 2
\& def close(h):
\&   # no return value
.Ve
.Sp
After \f(CW\*(C`close\*(C'\fR returns, the reference count of the handle is
decremented in the C part, which usually means that the handle and its
contents will be garbage collected.
.ie n .IP """export_description""" 4
.el .IP "\f(CWexport_description\fR" 4
.IX Item "export_description"
(Optional)
.Sp
.Vb 3
\& def export_description(h):
\&   # return a string
\&   return "description"
.Ve
.ie n .IP """get_size""" 4
.el .IP "\f(CWget_size\fR" 4
.IX Item "get_size"
(Required)
.Sp
.Vb 2
\& def get_size(h):
\&   # return the size of the disk
.Ve
.ie n .IP """is_rotational""" 4
.el .IP "\f(CWis_rotational\fR" 4
.IX Item "is_rotational"
(Optional)
.Sp
.Vb 2
\& def is_rotational(h):
\&   # return a boolean
.Ve
.ie n .IP """can_multi_conn""" 4
.el .IP "\f(CWcan_multi_conn\fR" 4
.IX Item "can_multi_conn"
(Optional)
.Sp
.Vb 2
\& def can_multi_conn(h):
\&   # return a boolean
.Ve
.ie n .IP """can_write""" 4
.el .IP "\f(CWcan_write\fR" 4
.IX Item "can_write"
(Optional)
.Sp
.Vb 2
\& def can_write(h):
\&   # return a boolean
.Ve
.ie n .IP """can_flush""" 4
.el .IP "\f(CWcan_flush\fR" 4
.IX Item "can_flush"
(Optional)
.Sp
.Vb 2
\& def can_flush(h):
\&   # return a boolean
.Ve
.ie n .IP """can_trim""" 4
.el .IP "\f(CWcan_trim\fR" 4
.IX Item "can_trim"
(Optional)
.Sp
.Vb 2
\& def can_trim(h):
\&   # return a boolean
.Ve
.ie n .IP """can_zero""" 4
.el .IP "\f(CWcan_zero\fR" 4
.IX Item "can_zero"
(Optional)
.Sp
.Vb 2
\& def can_zero(h):
\&   # return a boolean
.Ve
.ie n .IP """can_fast_zero""" 4
.el .IP "\f(CWcan_fast_zero\fR" 4
.IX Item "can_fast_zero"
(Optional)
.Sp
.Vb 2
\& def can_fast_zero(h):
\&   # return a boolean
.Ve
.ie n .IP """can_fua""" 4
.el .IP "\f(CWcan_fua\fR" 4
.IX Item "can_fua"
(Optional)
.Sp
.Vb 3
\& def can_fua(h):
\&   # return nbdkit.FUA_NONE or nbdkit.FUA_EMULATE
\&   # or nbdkit.FUA_NATIVE
.Ve
.ie n .IP """can_cache""" 4
.el .IP "\f(CWcan_cache\fR" 4
.IX Item "can_cache"
(Optional)
.Sp
.Vb 3
\& def can_cache(h):
\&   # return nbdkit.CACHE_NONE or nbdkit.CACHE_EMULATE
\&   # or nbdkit.CACHE_NATIVE
.Ve
.ie n .IP """can_extents""" 4
.el .IP "\f(CWcan_extents\fR" 4
.IX Item "can_extents"
(Optional)
.Sp
.Vb 2
\& def can_extents(h):
\&   # return a boolean
.Ve
.ie n .IP """pread""" 4
.el .IP "\f(CWpread\fR" 4
.IX Item "pread"
(Required)
.Sp
.Vb 2
\& def pread(h, buf, offset, flags):
\&   # read into the buffer
.Ve
.Sp
The body of your \f(CW\*(C`pread\*(C'\fR function should read exactly \f(CW\*(C`len(buf)\*(C'\fR
bytes of data starting at disk \f(CW\*(C`offset\*(C'\fR and write it into the buffer
\&\f(CW\*(C`buf\*(C'\fR.  \f(CW\*(C`flags\*(C'\fR is always 0.
.Sp
\&\s-1NBD\s0 only supports whole reads, so your function should try to read
the whole region (perhaps requiring a loop).  If the read fails or
is partial, your function should throw an exception, optionally using
\&\f(CW\*(C`nbdkit.set_error\*(C'\fR first.
.ie n .IP """pwrite""" 4
.el .IP "\f(CWpwrite\fR" 4
.IX Item "pwrite"
(Optional)
.Sp
.Vb 3
\& def pwrite(h, buf, offset, flags):
\&   length = len(buf)
\&   # no return value
.Ve
.Sp
The body of your \f(CW\*(C`pwrite\*(C'\fR function should write the buffer \f(CW\*(C`buf\*(C'\fR to
the disk.  You should write \f(CW\*(C`count\*(C'\fR bytes to the disk starting at
\&\f(CW\*(C`offset\*(C'\fR.  \f(CW\*(C`flags\*(C'\fR may contain \f(CW\*(C`nbdkit.FLAG_FUA\*(C'\fR.
.Sp
\&\s-1NBD\s0 only supports whole writes, so your function should try to
write the whole region (perhaps requiring a loop).  If the write
fails or is partial, your function should throw an exception,
 optionally using \f(CW\*(C`nbdkit.set_error\*(C'\fR first.
.ie n .IP """flush""" 4
.el .IP "\f(CWflush\fR" 4
.IX Item "flush"
(Optional)
.Sp
.Vb 2
\& def flush(h, flags):
\&   # no return value
.Ve
.Sp
The body of your \f(CW\*(C`flush\*(C'\fR function should do a \fBsync\fR\|(2) or
\&\fBfdatasync\fR\|(2) or equivalent on the backing store.
\&\f(CW\*(C`flags\*(C'\fR is always 0.
.Sp
If the flush fails, your function should throw an exception, optionally
using \f(CW\*(C`nbdkit.set_error\*(C'\fR first.
.ie n .IP """trim""" 4
.el .IP "\f(CWtrim\fR" 4
.IX Item "trim"
(Optional)
.Sp
.Vb 2
\& def trim(h, count, offset, flags):
\&   # no return value
.Ve
.Sp
The body of your \f(CW\*(C`trim\*(C'\fR function should \*(L"punch a hole\*(R" in the backing
store.  \f(CW\*(C`flags\*(C'\fR may contain \f(CW\*(C`nbdkit.FLAG_FUA\*(C'\fR.  If the trim fails,
your function should throw an exception, optionally using
\&\f(CW\*(C`nbdkit.set_error\*(C'\fR first.
.ie n .IP """zero""" 4
.el .IP "\f(CWzero\fR" 4
.IX Item "zero"
(Optional)
.Sp
.Vb 2
\& def zero(h, count, offset, flags):
\&   # no return value
.Ve
.Sp
The body of your \f(CW\*(C`zero\*(C'\fR function should ensure that \f(CW\*(C`count\*(C'\fR bytes of
the disk, starting at \f(CW\*(C`offset\*(C'\fR, will read back as zero.  \f(CW\*(C`flags\*(C'\fR is
a bitmask which may include \f(CW\*(C`nbdkit.FLAG_MAY_TRIM\*(C'\fR,
\&\f(CW\*(C`nbdkit.FLAG_FUA\*(C'\fR, \f(CW\*(C`nbdkit.FLAG_FAST_ZERO\*(C'\fR.
.Sp
\&\s-1NBD\s0 only supports whole writes, so your function should try to
write the whole region (perhaps requiring a loop).
.Sp
If the write fails or is partial, your function should throw an
exception, optionally using \f(CW\*(C`nbdkit.set_error\*(C'\fR first.  In particular,
if you would like to automatically fall back to \f(CW\*(C`pwrite\*(C'\fR (perhaps
because there is nothing to optimize if
\&\f(CW\*(C`flags & nbdkit.FLAG_MAY_TRIM\*(C'\fR is false), use
\&\f(CW\*(C`nbdkit.set_error(errno.EOPNOTSUPP)\*(C'\fR.
.ie n .IP """cache""" 4
.el .IP "\f(CWcache\fR" 4
.IX Item "cache"
(Optional)
.Sp
.Vb 2
\& def cache(h, count, offset, flags):
\&   # no return value
.Ve
.Sp
The body of your \f(CW\*(C`cache\*(C'\fR function should prefetch data in the
indicated range.
.Sp
If the cache operation fails, your function should throw an exception,
optionally using \f(CW\*(C`nbdkit.set_error\*(C'\fR first.
.ie n .IP """extents""" 4
.el .IP "\f(CWextents\fR" 4
.IX Item "extents"
(Optional)
.Sp
.Vb 4
\& def extents(h, count, offset, flags):
\&   # return an iterable object (eg. list) of
\&   # (offset, length, type) tuples:
\&   return [ (off1, len1, type1), (off2, len2, type2), ... ]
.Ve
.SS "Missing callbacks"
.IX Subsection "Missing callbacks"
.ie n .IP "Missing: ""load"" and ""unload""" 4
.el .IP "Missing: \f(CWload\fR and \f(CWunload\fR" 4
.IX Item "Missing: load and unload"
These are not needed because you can just use ordinary Python
constructs.
.ie n .IP "Missing: ""name"", ""version"", ""longname"", ""description"", ""config_help"", ""magic_config_key""." 4
.el .IP "Missing: \f(CWname\fR, \f(CWversion\fR, \f(CWlongname\fR, \f(CWdescription\fR, \f(CWconfig_help\fR, \f(CWmagic_config_key\fR." 4
.IX Item "Missing: name, version, longname, description, config_help, magic_config_key."
These are not yet supported.
.SS "Threads"
.IX Subsection "Threads"
The thread model for Python callbacks defaults to
\&\f(CW\*(C`NBDKIT_THREAD_MODEL_SERIALIZE_ALL_REQUESTS\*(C'\fR.  Since nbdkit 1.22
it is possible to set this by implementing a \f(CW\*(C`thread_model\*(C'\fR function
which returns one of the constants \f(CW\*(C`nbdkit.THREAD_MODEL_*\*(C'\fR.
.PP
The Python Global Interpreter Lock (\s-1GIL\s0) usually means that you cannot
execute Python code in parallel, but Python code which calls into
libraries which block (eg. to make \s-1HTTP\s0 requests) might be executed in
parallel.
.SH "FILES"
.IX Header "FILES"
.IP "\fI\f(CI$plugindir\fI/nbdkit\-python\-plugin.so\fR" 4
.IX Item "$plugindir/nbdkit-python-plugin.so"
The plugin.
.Sp
Use \f(CW\*(C`nbdkit \-\-dump\-config\*(C'\fR to find the location of \f(CW$plugindir\fR.
.SH "VERSION"
.IX Header "VERSION"
\&\f(CW\*(C`nbdkit\-python\-plugin\*(C'\fR first appeared in nbdkit 1.2.
.SH "SEE ALSO"
.IX Header "SEE ALSO"
\&\fBnbdkit\fR\|(1),
\&\fBnbdkit\-plugin\fR\|(3),
\&\fBpython\fR\|(1).
.SH "AUTHORS"
.IX Header "AUTHORS"
Eric Blake
.PP
Richard W.M. Jones
.PP
Nir Soffer
.SH "COPYRIGHT"
.IX Header "COPYRIGHT"
Copyright (C) 2013\-2020 Red Hat Inc.
.SH "LICENSE"
.IX Header "LICENSE"
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
.IP "\(bu" 4
Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
.IP "\(bu" 4
Redistributions in binary form must reproduce the above copyright
notice, this list of conditions and the following disclaimer in the
documentation and/or other materials provided with the distribution.
.IP "\(bu" 4
Neither the name of Red Hat nor the names of its contributors may be
used to endorse or promote products derived from this software without
specific prior written permission.
.PP
\&\s-1THIS SOFTWARE IS PROVIDED BY RED HAT AND CONTRIBUTORS\s0 ''\s-1AS IS\s0'' \s-1AND
ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,
THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL RED HAT OR
CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES\s0 (\s-1INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES\s0; \s-1LOSS OF
USE, DATA, OR PROFITS\s0; \s-1OR BUSINESS INTERRUPTION\s0) \s-1HOWEVER CAUSED AND
ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
OR TORT\s0 (\s-1INCLUDING NEGLIGENCE OR OTHERWISE\s0) \s-1ARISING IN ANY WAY OUT
OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
SUCH DAMAGE.\s0