Merge pull request #10548 from MaxNoe/pgf_pdf_pages

anntzer · web-flow · commit a2db8821a365 · 2018-02-28T01:10:15.000-08:00
Implement PdfPages for backend pgf
diff --git a/.gitignore b/.gitignore
@@ -80,6 +80,7 @@ result_images
 
 # Nose/Pytest generated files #
 ###############################
+.pytest_cache/
 .cache/
 .coverage
 .coverage.*
diff --git a/.travis.yml b/.travis.yml
@@ -37,6 +37,7 @@ addons:
       - texlive-latex-extra
       - texlive-latex-recommended
       - texlive-xetex
+      - texlive-luatex
 
 env:
   global:
diff --git a/doc/faq/howto_faq.rst b/doc/faq/howto_faq.rst
@@ -136,6 +136,10 @@ Finally, the multipage pdf object has to be closed::
 
     pp.close()
 
+The same can be done using the pgf backend::
+    
+    from matplotlib.backends.backend_pgf import PdfPages
+
 
 .. _howto-subplots-adjust:
 
diff --git a/doc/users/next_whats_new/pgf_pdfpages.rst b/doc/users/next_whats_new/pgf_pdfpages.rst
@@ -0,0 +1,19 @@
+Multipage PDF support for pgf backend
+-------------------------------------
+
+The pgf backend now also supports multipage PDF files.
+
+.. code-block:: python
+
+    from matplotlib.backends.backend_pgf import PdfPages
+    import matplotlib.pyplot as plt
+
+    with PdfPages('multipage.pdf') as pdf:
+        # page 1
+        plt.plot([2, 1, 3])
+        pdf.savefig()
+
+        # page 2
+        plt.cla()
+        plt.plot([3, 1, 2])
+        pdf.savefig()
diff --git a/doc/users/whats_new.rst b/doc/users/whats_new.rst
@@ -14,12 +14,12 @@ revision, see the :ref:`github-stats`.
 ..
    For a release, add a new section after this, then comment out the include
    and toctree below by indenting them. Uncomment them after the release.
-   .. include:: next_whats_new/README.rst
-   .. toctree::
-      :glob:
-      :maxdepth: 1
+.. include:: next_whats_new/README.rst
+.. toctree::
+   :glob:
+   :maxdepth: 1
 
-      next_whats_new/*
+   next_whats_new/*
 
 
 New in Matplotlib 2.2
diff --git a/examples/misc/multipage_pdf.py b/examples/misc/multipage_pdf.py
@@ -5,6 +5,10 @@
 
 This is a demo of creating a pdf file with several pages,
 as well as adding metadata and annotations to pdf files.
+
+If you want to use a multipage pdf file using LaTeX, you need
+to use `from matplotlib.backends.backend_pgf import PdfPages`.
+This version however does not support `attach_note`.
 """
 
 import datetime
diff --git a/lib/matplotlib/backends/backend_pgf.py b/lib/matplotlib/backends/backend_pgf.py
@@ -17,13 +17,15 @@
 import weakref
 
 import matplotlib as mpl
-from matplotlib import _png, rcParams
+from matplotlib import _png, rcParams, __version__
 from matplotlib.backend_bases import (
     _Backend, FigureCanvasBase, FigureManagerBase, GraphicsContextBase,
     RendererBase)
 from matplotlib.backends.backend_mixed import MixedModeRenderer
 from matplotlib.cbook import is_writable_file_like
 from matplotlib.path import Path
+from matplotlib.figure import Figure
+from matplotlib._pylab_helpers import Gcf
 
 
 ###############################################################################
@@ -50,13 +52,30 @@
         warnings.warn('error getting fonts from fc-list', UserWarning)
 
 
+_luatex_version_re = re.compile(
+    'This is LuaTeX, Version (?:beta-)?([0-9]+)\.([0-9]+)\.([0-9]+)'
+)
+
+
 def get_texcommand():
     """Get chosen TeX system from rc."""
     texsystem_options = ["xelatex", "lualatex", "pdflatex"]
     texsystem = rcParams["pgf.texsystem"]
     return texsystem if texsystem in texsystem_options else "xelatex"
 
 
+def _get_lualatex_version():
+    """Get version of luatex"""
+    output = subprocess.check_output(['lualatex', '--version'])
+    return _parse_lualatex_version(output.decode())
+
+
+def _parse_lualatex_version(output):
+    '''parse the lualatex version from the output of `lualatex --version`'''
+    match = _luatex_version_re.match(output)
+    return tuple(map(int, match.groups()))
+
+
 def get_fontspec():
     """Build fontspec preamble from rc."""
     latex_fontspec = []
@@ -990,4 +1009,221 @@ def _cleanup_all():
     LatexManager._cleanup_remaining_instances()
     TmpDirCleaner.cleanup_remaining_tmpdirs()
 
+
 atexit.register(_cleanup_all)
+
+
+class PdfPages:
+    """
+    A multi-page PDF file using the pgf backend
+
+    Examples
+    --------
+
+    >>> import matplotlib.pyplot as plt
+    >>> # Initialize:
+    >>> with PdfPages('foo.pdf') as pdf:
+    ...     # As many times as you like, create a figure fig and save it:
+    ...     fig = plt.figure()
+    ...     pdf.savefig(fig)
+    ...     # When no figure is specified the current figure is saved
+    ...     pdf.savefig()
+    """
+    __slots__ = (
+        '_outputfile',
+        'keep_empty',
+        '_tmpdir',
+        '_basename',
+        '_fname_tex',
+        '_fname_pdf',
+        '_n_figures',
+        '_file',
+        'metadata',
+    )
+
+    def __init__(self, filename, *, keep_empty=True, metadata=None):
+        """
+        Create a new PdfPages object.
+
+        Parameters
+        ----------
+
+        filename : str
+            Plots using :meth:`PdfPages.savefig` will be written to a file at
+            this location. Any older file with the same name is overwritten.
+        keep_empty : bool, optional
+            If set to False, then empty pdf files will be deleted automatically
+            when closed.
+        metadata : dictionary, optional
+            Information dictionary object (see PDF reference section 10.2.1
+            'Document Information Dictionary'), e.g.:
+            `{'Creator': 'My software', 'Author': 'Me',
+            'Title': 'Awesome fig'}`
+
+            The standard keys are `'Title'`, `'Author'`, `'Subject'`,
+            `'Keywords'`, `'Producer'`, `'Creator'` and `'Trapped'`.
+            Values have been predefined for `'Creator'` and `'Producer'`.
+            They can be removed by setting them to the empty string.
+        """
+        self._outputfile = filename
+        self._n_figures = 0
+        self.keep_empty = keep_empty
+        self.metadata = metadata or {}
+
+        # create temporary directory for compiling the figure
+        self._tmpdir = tempfile.mkdtemp(prefix="mpl_pgf_pdfpages_")
+        self._basename = 'pdf_pages'
+        self._fname_tex = os.path.join(self._tmpdir, self._basename + ".tex")
+        self._fname_pdf = os.path.join(self._tmpdir, self._basename + ".pdf")
+        self._file = open(self._fname_tex, 'wb')
+
+    def _write_header(self, width_inches, height_inches):
+        supported_keys = {
+            'title', 'author', 'subject', 'keywords', 'creator',
+            'producer', 'trapped'
+        }
+        infoDict = {
+            'creator': 'matplotlib %s, https://matplotlib.org' % __version__,
+            'producer': 'matplotlib pgf backend %s' % __version__,
+        }
+        metadata = {k.lower(): v for k, v in self.metadata.items()}
+        infoDict.update(metadata)
+        hyperref_options = ''
+        for k, v in infoDict.items():
+            if k not in supported_keys:
+                raise ValueError(
+                    'Not a supported pdf metadata field: "{}"'.format(k)
+                )
+            hyperref_options += 'pdf' + k + '={' + str(v) + '},'
+
+        latex_preamble = get_preamble()
+        latex_fontspec = get_fontspec()
+        latex_header = r"""\PassOptionsToPackage{{
+  {metadata}
+}}{{hyperref}}
+\RequirePackage{{hyperref}}
+\documentclass[12pt]{{minimal}}
+\usepackage[
+    paperwidth={width}in,
+    paperheight={height}in,
+    margin=0in
+]{{geometry}}
+{preamble}
+{fontspec}
+\usepackage{{pgf}}
+\setlength{{\parindent}}{{0pt}}
+
+\begin{{document}}%%
+""".format(
+            width=width_inches,
+            height=height_inches,
+            preamble=latex_preamble,
+            fontspec=latex_fontspec,
+            metadata=hyperref_options,
+        )
+        self._file.write(latex_header.encode('utf-8'))
+
+    def __enter__(self):
+        return self
+
+    def __exit__(self, exc_type, exc_val, exc_tb):
+        self.close()
+
+    def close(self):
+        """
+        Finalize this object, running LaTeX in a temporary directory
+        and moving the final pdf file to `filename`.
+        """
+        self._file.write(rb'\end{document}\n')
+        self._file.close()
+
+        if self._n_figures > 0:
+            try:
+                self._run_latex()
+            finally:
+                try:
+                    shutil.rmtree(self._tmpdir)
+                except:
+                    TmpDirCleaner.add(self._tmpdir)
+        elif self.keep_empty:
+            open(self._outputfile, 'wb').close()
+
+    def _run_latex(self):
+        texcommand = get_texcommand()
+        cmdargs = [
+            str(texcommand),
+            "-interaction=nonstopmode",
+            "-halt-on-error",
+            os.path.basename(self._fname_tex),
+        ]
+        try:
+            subprocess.check_output(
+                cmdargs, stderr=subprocess.STDOUT, cwd=self._tmpdir
+            )
+        except subprocess.CalledProcessError as e:
+            raise RuntimeError(
+                "%s was not able to process your file.\n\nFull log:\n%s"
+                % (texcommand, e.output.decode('utf-8')))
+
+        # copy file contents to target
+        shutil.copyfile(self._fname_pdf, self._outputfile)
+
+    def savefig(self, figure=None, **kwargs):
+        """
+        Saves a :class:`~matplotlib.figure.Figure` to this file as a new page.
+
+        Any other keyword arguments are passed to
+        :meth:`~matplotlib.figure.Figure.savefig`.
+
+        Parameters
+        ----------
+
+        figure : :class:`~matplotlib.figure.Figure` or int, optional
+            Specifies what figure is saved to file. If not specified, the
+            active figure is saved. If a :class:`~matplotlib.figure.Figure`
+            instance is provided, this figure is saved. If an int is specified,
+            the figure instance to save is looked up by number.
+        """
+        if not isinstance(figure, Figure):
+            if figure is None:
+                manager = Gcf.get_active()
+            else:
+                manager = Gcf.get_fig_manager(figure)
+            if manager is None:
+                raise ValueError("No figure {}".format(figure))
+            figure = manager.canvas.figure
+
+        try:
+            orig_canvas = figure.canvas
+            figure.canvas = FigureCanvasPgf(figure)
+
+            width, height = figure.get_size_inches()
+            if self._n_figures == 0:
+                self._write_header(width, height)
+            else:
+                self._file.write(self._build_newpage_command(width, height))
+
+            figure.savefig(self._file, format="pgf", **kwargs)
+            self._n_figures += 1
+        finally:
+            figure.canvas = orig_canvas
+
+    def _build_newpage_command(self, width, height):
+        '''LuaLaTeX from version 0.85 removed the `\pdf*` primitives,
+        so we need to check the lualatex version and use `\pagewidth` if
+        the version is 0.85 or newer
+        '''
+        texcommand = get_texcommand()
+        if texcommand == 'lualatex' and _get_lualatex_version() >= (0, 85, 0):
+            cmd = r'\page'
+        else:
+            cmd = r'\pdfpage'
+
+        newpage = r'\newpage{cmd}width={w}in,{cmd}height={h}in%' + '\n'
+        return newpage.format(cmd=cmd, w=width, h=height).encode('utf-8')
+
+    def get_pagecount(self):
+        """
+        Returns the current number of pages in the multipage pdf file.
+        """
+        return self._n_figures
diff --git a/lib/matplotlib/tests/test_backend_pgf.py b/lib/matplotlib/tests/test_backend_pgf.py
diff --git a/tutorials/text/pgf.py b/tutorials/text/pgf.py
