Merge pull request getpelican#671 from wilsonfreitas/rmd

rmd_reader: New features and py3 compatibility

rmd_reader/Readme.md

@@ -41,28 +41,26 @@ This plugin calls R to process these files and generates markdown files that are
 
 ### Configuration
 
-`rmd_reader` has these 3 variables that can be set in `pelicanconf`.
+`rmd_reader` has these variables that can be set in `pelicanconf`.
 
 - `RMD_READER_CLEANUP` (`True`): The RMarkdown file is converted into a Markdown file with the extension `.aux` (to avoid conflicts while pelican is processing). This file is processed by pelican's MarkdownReader and is removed after that (the cleanup step). So if you want to check this file set `RMD_READER_CLEANUP=True`.
+- `RMD_READER_RENAME_PLOT` (`True`): the figures generated for plots are named with a default prefix (usually `unnamed-chunk`) followed by a sequential number. That sequence starts on 1 for every processed file, which causes naming conflicts among files. In order to avoid these conflicts `RMD_READER_RENAME_PLOT` can be set `True` and that prefix is replaced with the same name of the post file, without extension. Another way to avoid conflicts is naming the chuncks and in that case this variable can be set `False`.
 - `RMD_READER_KNITR_QUIET` (`True`): sets `knitr`'s quiet argument.
 - `RMD_READER_KNITR_ENCODING` (`UTF-8`): sets `knitr`'s encoding argument.
 - `RMD_READER_KNITR_OPTS_CHUNK` (`None`): sets `knitr`'s `opts_chunk`.
+- `RMD_READER_KNITR_OPTS_KNIT` (`None`): sets `knitr`'s `opts_knit`.
 
 
 ### Plotting
 
-The code below must be pasted inside the `.Rmd` file in order to correctly set the `src` attribute of `img` tag.
-
-	```{r, echo=FALSE}
-	hook_plot <- knit_hooks$get('plot')
-	knit_hooks$set(plot=function(x, options) {
-	    if (!is.null(options$pelican.publish) && options$pelican.publish) {
-	        x <- paste0("{filename}", x)
-	    }
-	    hook_plot(x, options)
-	})
-	opts_chunk$set(pelican.publish=TRUE)
-	```
-
-I usually paste it just after the Markdown header.
-There is a R [template](https://github.com/almartin82/pelicanRMD) available that has the base elements needed by `rmd_reader`.
+I strongly suggest using the variable `RMD_READER_RENAME_PLOT=True`.
+That helps with avoiding naming conflits among different posts.
+`rmd_reader` sets knitr's `unnamed.chunk.label` option to the Rmd file name (without extension) in runtime.
+
+Alternatively, Rebecca Weiss (@rjweiss) suggested using `opts_chunk` to set knitr's `fig.path` ([link](http://rjweiss.github.io/articles/2014_08_25/testing-rmarkdown-integration/)).
+Now that can be done directly in `pelicanconf` thougth `RMD_READER_KNITR_OPTS_CHUNK`, that variable receives a `dict` with options to be passed to knitr's `opts_chunk`.
+
+```
+RMD_READER_KNITR_OPTS_CHUNK = {'fig.path': '../../../figure/'}
+```
+

rmd_reader/rmd_reader.py

@@ -4,32 +4,34 @@
 import warnings
 import logging
 
-logger = logging.getLogger(__name__)
+logger = logging.getLogger('RMD_READER')
 
 from pelican import readers
 from pelican import signals
 from pelican import settings
-from pelican.utils import pelican_open
-from markdown import Markdown
 
 knitr = None
 rmd = False
 
 def initsignal(pelicanobj):
-    global knitr,rmd
+    global knitr, rmd, robjects
     try:
         with warnings.catch_warnings():
             warnings.simplefilter("ignore")
             from rpy2.robjects.packages import importr
+            import rpy2.robjects as robjects
         knitr = importr('knitr')
         idx = knitr.opts_knit.names.index('set')
         PATH = pelicanobj.settings.get('PATH','%s/content' % settings.DEFAULT_CONFIG.get('PATH'))
         logger.debug("RMD_READER PATH = %s", PATH)
         knitr.opts_knit[idx](**{'base.dir': PATH})
+        knitroptsknit = pelicanobj.settings.get('RMD_READER_KNITR_OPTS_KNIT', None)
+        if knitroptsknit:
+            knitr.opts_knit[idx](**{str(k): v for k,v in knitroptsknit.items()})
         idx = knitr.opts_chunk.names.index('set')
         knitroptschunk = pelicanobj.settings.get('RMD_READER_KNITR_OPTS_CHUNK', None)
-        if knitroptschunk is not None:     
-            knitr.opts_chunk[idx](**{str(k): v for k,v in knitroptschunk.iteritems()})
+        if knitroptschunk:
+            knitr.opts_chunk[idx](**{str(k): v for k,v in knitroptschunk.items()})
         rmd = True
     except ImportError as ex:
         rmd = False
@@ -39,24 +41,33 @@ class RmdReader(readers.BaseReader):
 
     @property
     def enabled():
-        global rmd
         return rmd
 
     # You need to have a read method, which takes a filename and returns
     # some content and the associated metadata.
     def read(self, filename):
         """Parse content and metadata of markdown files"""
-        global knitr
         QUIET = self.settings.get('RMD_READER_KNITR_QUIET', True)
         ENCODING = self.settings.get('RMD_READER_KNITR_ENCODING', 'UTF-8')
         CLEANUP = self.settings.get('RMD_READER_CLEANUP', True)
+        RENAME_PLOT = self.settings.get('RMD_READER_RENAME_PLOT', True)
         logger.debug("RMD_READER_KNITR_QUIET = %s", QUIET)
-        logger.debug("RMD_READER_KNITR_QUIET = %s", ENCODING)
+        logger.debug("RMD_READER_KNITR_ENCODING = %s", ENCODING)
         logger.debug("RMD_READER_CLEANUP = %s", CLEANUP)
+        logger.debug("RMD_READER_RENAME_PLOT = %s", RENAME_PLOT)
         # replace single backslashes with double backslashes
         filename = filename.replace('\\', '\\\\')
         # parse Rmd file - generate md file
         md_filename = filename.replace('.Rmd', '.aux').replace('.rmd', '.aux')
+        if RENAME_PLOT:
+            chunk_label = os.path.splitext(os.path.basename(filename))[0]
+            logger.debug('Chunk label: %s', chunk_label)
+            robjects.r('''
+opts_knit$set(unnamed.chunk.label="{unnamed_chunk_label}")
+render_markdown()
+hook_plot <- knit_hooks$get('plot')
+knit_hooks$set(plot=function(x, options) hook_plot(paste0("{{filename}}/", x), options))
+            '''.format(unnamed_chunk_label=chunk_label))
         knitr.knit(filename, md_filename, quiet=QUIET, encoding=ENCODING)
         # read md file - create a MarkdownReader
         md_reader = readers.MarkdownReader(self.settings)

rmd_reader/test_rmd_reader.py

@@ -10,7 +10,7 @@
 from pelican import Pelican
 from pelican.settings import read_settings
 
-logging.basicConfig(stream=sys.stderr, level=logging.CRITICAL)
+logging.basicConfig(stream=sys.stderr, level=logging.DEBUG)
 
 class Test(unittest.TestCase):
 
@@ -19,19 +19,23 @@ def setUp(self):
         try:
             import rpy2
             import rmd_reader
-        except Exception, e:
+        except Exception:
             raise unittest.SkipTest("rpy not installed.  Will not test rmd_reader.")
 
         self.testtitle = 'rtest'
         self.cwd = os.path.dirname(os.path.abspath(__file__))
+        logging.debug(self.cwd)
 
         # Setup content dir and test rmd file
         self.contentdir = os.path.join(self.cwd,'test-content')
+        logging.debug(self.contentdir)
         try:
             os.mkdir(self.contentdir)
         except Exception:
             pass
         self.contentfile = os.path.join(self.contentdir,'test.rmd')
+        logging.debug(self.contentfile)
+
         self.testrmd = '''Title: %s
 Date: 2014-06-23
 
@@ -46,6 +50,8 @@ def setUp(self):
 
         # Setup output dir
         self.outputdir = os.path.join(self.cwd,'test-output')
+        logging.debug(self.outputdir)
+
         try:
             os.mkdir(self.outputdir)
         except Exception:
@@ -56,17 +62,20 @@ def setUp(self):
 
 
     def tearDown(self):
+        logging.debug('CLEAN')
         if os.path.isdir(self.outputdir):
             shutil.rmtree(self.outputdir)
         if os.path.isdir(self.contentdir):
             shutil.rmtree(self.contentdir)
 
-
     def testKnitrSettings(self):
         settings = read_settings(path=None, override={
+            'LOAD_CONTENT_CACHE': False,
             'PATH': self.contentdir,
             'OUTPUT_PATH': self.outputdir,
-            'KNITR_OPTS_CHUNK': {'fig.path' : '%s/' % self.figpath},
+            'RMD_READER_KNITR_OPTS_CHUNK': {'fig.path' : '%s/' % self.figpath},
+            'RMD_READER_KNITR_OPTS_KNIT': {'progress' : True, 'verbose': True},
+            'RMD_READER_RENAME_PLOT': False,
             'PLUGIN_PATHS': ['../'],
             'PLUGINS': ['rmd_reader'],
         })
@@ -75,12 +84,48 @@ def testKnitrSettings(self):
 
         outputfilename = os.path.join(self.outputdir,'%s.html' % self.testtitle)
         self.assertTrue(os.path.exists(outputfilename),'File %s was not created.' % outputfilename)
-        imagesdir = os.path.join(self.outputdir,self.figpath)
+
+        imagesdir = os.path.join(self.outputdir, self.figpath)
         self.assertTrue(os.path.exists(imagesdir), 'figpath not created.')
+
+        imagefile = os.path.join(imagesdir, 'unnamed-chunk') + '-1-1.png'
+        logging.debug(imagefile)
         images = glob.glob('%s/*' % imagesdir)
+        logging.debug(images)
+        self.assertTrue(os.path.exists(imagefile), 'image correctly named.')
+
         self.assertTrue(len(images) == 1,'Contents of images dir is not correct: %s' % ','.join(images))
 
 
+    def testKnitrSettings2(self):
+        settings = read_settings(path=None, override={
+            'LOAD_CONTENT_CACHE': False,
+            'PATH': self.contentdir,
+            'OUTPUT_PATH': self.outputdir,
+            'RMD_READER_KNITR_OPTS_CHUNK': {'fig.path' : '%s/' % self.figpath},
+            'RMD_READER_KNITR_OPTS_KNIT': {'progress' : True, 'verbose': True},
+            'RMD_READER_RENAME_PLOT': True,
+            'PLUGIN_PATHS': ['../'],
+            'PLUGINS': ['rmd_reader'],
+        })
+        pelican = Pelican(settings=settings)
+        pelican.run()
+
+        outputfilename = os.path.join(self.outputdir,'%s.html' % self.testtitle)
+        self.assertTrue(os.path.exists(outputfilename),'File %s was not created.' % outputfilename)
+
+        imagesdir = os.path.join(self.outputdir, self.figpath)
+        self.assertTrue(os.path.exists(imagesdir), 'figpath not created.')
+
+        imagefile = os.path.join(imagesdir, os.path.splitext(os.path.split(self.contentfile)[1])[0]) + '-1-1.png'
+        logging.debug(imagefile)
+        self.assertTrue(os.path.exists(imagefile), 'image correctly named.')
+
+        images = glob.glob('%s/*' % imagesdir)
+        logging.debug(images)
+        self.assertTrue(len(images) == 1,'Contents of images dir is not correct: %s' % ','.join(images))
+
+
 
 if __name__ == "__main__":
     #import sys;sys.argv = ['', 'Test.testName']