Updates copyres so files are gathered upfront and available in templa…

…tes as "respaths" and "resfilenames"

Also creates change events for hooks.
Change sitemap to exclude resources (can add via config SITEMAP_RESOURCE_EXT, excluding images/video)

acrylamid/readers.py

@@ -12,10 +12,11 @@
 import abc
 import codecs
 import traceback
+import glob
 
 BOM_UTF8 = codecs.BOM_UTF8.decode('utf8')
 
-from os.path import join, getmtime, relpath, splitext
+from os.path import join, getmtime, relpath, splitext, dirname, basename, isfile, normpath
 from fnmatch import fnmatch
 from datetime import datetime, tzinfo, timedelta
 
@@ -294,6 +295,10 @@ def __init__(self, path, conf):
             if m.group(1) is not None:
                 meta['cats'] = m.group(1).split('/')
 
+        if 'copyres' in meta:
+            meta['respaths'] = self.getresources(meta.get('copyres'), conf)
+            meta['resfilenames'] = [basename(val) for val in meta['respaths']] 
+
         self.offset = i
         Reader.__init__(self, conf, meta)
 
@@ -322,6 +327,22 @@ def date(self):
         "Fallback to last modification timestamp if date is unset."
         return Date.fromtimestamp(getmtime(self.filename)).replace(tzinfo=self.tzinfo)
 
+    def getresources(self, wildcards, conf):
+        """Generate a list of resources files based on the wildcard(s) passed in."""
+        reslist = []
+        if isinstance(wildcards, list):
+            for term in wildcards:
+                # exclude missing and non file types 
+                reslist.extend([ normpath(f) for f in glob.glob(join(dirname(self.filename), term)) if isfile(f) ])
+        else:
+            if wildcards is None:
+                # use default wildcard appended to entry filename
+                reslist = [ normpath(f) for f in glob.glob(splitext(self.filename)[0] + conf.get('copyres_wildcard', '_[0-9]*.*')) if isfile(f) ]
+            else:
+                # provided wildcard appended to input directory
+                reslist = [ normpath(f) for f in glob.glob(join(dirname(self.filename), wildcards)) if isfile(f) ]
+        return reslist
+
 
 class MetadataMixin(object):
 
@@ -405,6 +426,14 @@ def tags(self):
             return [fx]
         return fx
 
+    @property
+    def respaths(self):
+        return self.props.get('respaths', [])
+
+    @property
+    def resfilenames(self):
+        return self.props.get('resfilenames', [])
+
     @property
     def draft(self):
         """If set to True, the entry will not appear in articles, index, feed and tag view."""

acrylamid/views/entry.py

@@ -8,15 +8,15 @@
 import io
 import glob
 
-from os.path import isfile, splitext, dirname, split, normpath
+from os.path import isfile, dirname, basename, getmtime, join
 from collections import defaultdict
 
 from acrylamid import refs
 from acrylamid.refs import modified, references
 from acrylamid.views import View
 from acrylamid import log
 from acrylamid.errors import AcrylamidException
-from acrylamid.helpers import expand, union, joinurl, event, link
+from acrylamid.helpers import expand, union, joinurl, event, link, mkfile
 
 
 class Base(View):
@@ -66,34 +66,34 @@ def generate(self, conf, env, data):
 
             if all([isfile(path), unmodified, not tt.modified, not entry.modified,
             not modified(*references(entry))]):
-                event.skip(self.name, path)
-                continue
-
-            html = tt.render(conf=conf, entry=entry, env=union(env,
-                             entrylist=[entry], type=self.__class__.__name__.lower(),
-                             prev=prev, next=next, route=expand(self.path, entry)))
-
-            yield html, path
-
-            if entry.hasproperty('copyres'):
-                copyres = entry.props.get('copyres')
-                if isinstance(copyres, list):
-                    copyreslist = [normpath(split(entry.filename)[0] + '/' + val) for val in copyres]
-                else:
-                    if copyres is None:
-                        # use default wildcard appended to entry filename
-                        wcard = splitext(entry.filename)[0] + conf.get('copyres_wildcard', '_[0-9]*.*')
-                    else:  
-                        # provided wildcard appended to input directory
-                        wcard = split(entry.filename)[0] + '/' + copyres
-                    copyreslist = glob.glob(wcard)
-
-                for resource in copyreslist:
+                entryskipped = True # log later incase a resource is updated, change will be reflected in sitemap
+            else:
+                entryskipped = False
+                html = tt.render(conf=conf, entry=entry, env=union(env,
+                                 entrylist=[entry], type=self.__class__.__name__.lower(),
+                                 prev=prev, next=next, route=expand(self.path, entry)))
+                yield html, path
+
+            # check if any resources need to be moved
+            if entry.hasproperty('respaths'):
+                for res_src in entry.props.get('respaths'):
+                    res_dest = join(dirname(path), basename(res_src))
+                    # Note, presence of res_src check in FileReader.getresources
+                    if isfile(res_dest) and getmtime(res_dest) > getmtime(res_src):
+                        event.skip(self.name, res_dest)
+                        continue
                     try:
-                        fp = io.open(resource, 'rb')
-                        yield fp , dirname(path) + '/' + split(resource)[1]
+                        fp = io.open(res_src, 'rb')
+                        if entryskipped == True:
+                            event.update(self.name, path)
+                            entryskipped = False
+                        # use mkfile rather than yield so different ns can be specified (and filtered by sitemap)
+                        mkfile(fp, res_dest, ns='resource', force=env.options.force, dryrun=env.options.dryrun)
                     except IOError as e:
-                        log.warn("Failed to copy '%s' whilst processing '%s' (%s)" % (resource, entry.filename, e.strerror))
+                        log.warn("Failed to copy resource '%s' whilst processing '%s' (%s)" % (res_src, entry.filename, e.strerror))
+            if entryskipped == True:
+                event.skip(self.name, path)
+
 
 class Entry(Base):
     """Creates single full-length entry

acrylamid/views/sitemap.py

@@ -6,7 +6,7 @@
 import io
 
 from time import strftime, gmtime
-from os.path import getmtime, exists
+from os.path import getmtime, exists, splitext
 
 from acrylamid.views import View
 from acrylamid.helpers import event, joinurl, rchop
@@ -53,14 +53,25 @@ class Sitemap(View):
     def init(self, conf, env):
 
         def track(ns, path):
-            self.files.add((ns, path))
+            if ns != 'resource':
+                self.files.add((ns, path))                
+            elif self.resext and splitext(path) in self.resext:
+                self.files.add((ns, path))
 
         def changed(ns, path):
             if not self.modified:
                 self.modified = True
 
         self.files = set([])
         self.modified = False
+
+        # use extension to check if resource should be tracked (eg PDF but should excluding images and video)
+        self.resext = conf.get('sitemap_resource_ext', [])
+        # Considered adding images and video, but need to be associated with loc. Better to have access to entry obj
+        # see http://support.google.com/webmasters/bin/answer.py?hl=en&answer=183668
+        #
+        #self.imgext = conf.get('sitemap_image_ext', [])
+        #self.vidext = conf.get('sitemap_video_ext', [])
 
         # track output files
         event.register(track, to=['create', 'update', 'skip', 'identical'])