Modified the download_data script to download the TIGER blocks.

[dead/census-tools.git] / src / Data.py

diff --git a/src/Data.py b/src/Data.py
index a8d962d1c52518fccfb881f2c1ff46dc8d099ba2..e08f4666788478991057996776d0ff051c6f8739 100644 (file)
--- a/src/Data.py
+++ b/src/Data.py
@@ -29,10 +29,23 @@ class State:
         self.counties = []
 
 
-    def tiger_data_url(self):
-        tdu =  self.TIGER_ROOT + '/'
-        tdu += str(self.id) + '_' + self.name.upper().replace(' ', '_')
-        return tdu
+    def tiger_data_root(self):
+        tdr =  self.TIGER_ROOT + '/'
+        tdr += str(self.id) + '_' + self.name.upper().replace(' ', '_')
+        return tdr
+
+
+    def blocks_data_url(self):
+        bdu =  self.tiger_data_root() + '/'
+        bdu += self.blocks_zipfile_name()
+        return bdu
+
+
+    def blocks_data_path(self):
+        bdp =  'data/census2000/'
+        bdp += self.name.lower().replace(' ', '_')
+        bdp += '/blocks'
+        return bdp
 
 
     def lines_data_path(self):
@@ -42,16 +55,29 @@ class State:
         return ldp
 
 
+    def blocks_zipfile_name(self):
+        return 'tl_2009_' + str(self.id) + '_tabblock00.zip'
+
+
+    def blocks_shapefile_path(self):
+        bsp =  self.blocks_data_path() + '/'
+        bsp += 'tl_2009_' + str(self.id) + '_tabblock00.shp'
+        return bsp
+
+
     def add_county(self, county_id, county_name, override_name=False):
         """
         We would like each county to have a pointer to its containing
         state. This so we can compute the file URL, directory, and so
         forth from within the county.
         """
-        self.counties.append(County(county_id, county_name, self, override_name))
+        self.counties.append(County(county_id,
+                                    county_name,
+                                    self,
+                                    override_name))
+
+
 
-        
-    
 class County:
     """
     A county represents either a county or city. It doesn't make
@@ -91,25 +117,58 @@ class County:
             # you and don't add the word 'County' on to it."
             return self.name
 
-    
-    def tiger_data_url(self):
-        tdp =  self.state.tiger_data_url() + '/'
+
+    def lines_data_url(self):
+        tdp =  self.state.tiger_data_root() + '/'
         tdp += self.state_county_id()
         tdp += '_' + self.full_name().replace(' ', '_') + '/'
-        tdp += self.zipfile_name()
+        tdp += self.lines_zipfile_name()
         return tdp
 
 
-    def zipfile_name(self):
+    def lines_zipfile_name(self):
         return 'tl_2009_' + self.state_county_id() + '_edges.zip'
 
-    
-    def shapefile_path(self):
+
+    def lines_shapefile_path(self):
         sfp =  self.state.lines_data_path() + '/'
         sfp += 'tl_2009_' + self.state_county_id() + '_edges.shp'
         return sfp
 
 
+
+def download_blocks(states):
+    """
+    Download the TIGER/Line block files for each state.
+    """
+
+    for state in states:
+        # First, create the blocks data path if it doesn't exist.
+        FileUtils.mkdir_p(state.blocks_data_path(), 0755)
+
+        if not os.path.exists(state.blocks_shapefile_path()):
+            url = state.blocks_data_url()
+            tmpfile = state.blocks_zipfile_name()
+            print "Grabbing data for %s." % state.name
+            print "Downloading %s to %s..." % (url, tmpfile)
+
+            try:
+                # This can fail for a bunch of reasons...
+                urllib.urlretrieve(url, tmpfile)
+                print "Unzipping %s to %s..." % (tmpfile, state.blocks_data_path())
+                z = zipfile.ZipFile(tmpfile)
+                z.extractall(state.blocks_data_path())
+            except:
+                # That we don't care about.
+                pass
+            finally:
+                # But we always clean up after ourselves.
+                print "Removing %s..." % tmpfile
+                FileUtils.rm_f(tmpfile)
+                print "Done.\n"
+
+
+
 def download_lines(states):
     """
     Download the TIGER/Line 'all lines' files for each county in states.
@@ -122,9 +181,9 @@ def download_lines(states):
         # Now loop through the counties, and download/unzip the lines
         # data if necessary.
         for county in state.counties:
-            if not os.path.exists(county.shapefile_path()):
+            if not os.path.exists(county.lines_shapefile_path()):
                 url = county.tiger_data_url()
-                tmpfile = county.zipfile_name()
+                tmpfile = county.lines_zipfile_name()
                 print "Grabbing data for %s (%s)." % (county.full_name(), state.name)
                 print "Downloading %s to %s..." % (url, tmpfile)
                 
@@ -138,7 +197,7 @@ def download_lines(states):
                     # That we don't care about.
                     pass
                 finally:
-                    # But we always clean up after ourselves.
+                     # But we always clean up after ourselves.
                     print "Removing %s..." % tmpfile
                     FileUtils.rm_f(tmpfile)
                     print "Done.\n"