X-Git-Url: http://gitweb.michael.orlitzky.com/?a=blobdiff_plain;f=makefile;h=b47b5f20252c06a8264aac783393c76c6bc324ff;hb=231be924cc841fd17af1d322823e95bab0967563;hp=432a515ded626f70a6102692e50fdf7731e7a912;hpb=8b5fdd58d539d5ef24a5e726738c39637d1a38ab;p=dead%2Fcensus-tools.git diff --git a/makefile b/makefile index 432a515..b47b5f2 100644 --- a/makefile +++ b/makefile @@ -1,23 +1,8 @@ DB_NAME=census DB_USER=postgres TIGER_SRID=4269 +SHAPELY_URL=http://pypi.python.org/packages/source/S/Shapely/Shapely-1.0.14.tar.gz -# Root folder for the shapefiles. -TIGER_ROOT=http://www2.census.gov/geo/tiger/TIGER2009 - -# State-specific folders. -DC_ROOT=$(TIGER_ROOT)/11_DISTRICT_OF_COLUMBIA -MD_ROOT=$(TIGER_ROOT)/24_MARYLAND -VA_ROOT=$(TIGER_ROOT)/51_VIRGINIA -PA_ROOT=$(TIGER_ROOT)/42_PENNSYLVANIA -NY_ROOT=$(TIGER_ROOT)/36_NEW_YORK - -# URLs for the TIGER/Line block-level shapefiles. -DC_BLOCKS_URL=$(DC_ROOT)/tl_2009_11_tabblock00.zip -MD_BLOCKS_URL=$(MD_ROOT)/tl_2009_24_tabblock00.zip -VA_BLOCKS_URL=$(VA_ROOT)/tl_2009_51_tabblock00.zip -PA_BLOCKS_URL=$(PA_ROOT)/tl_2009_42_tabblock00.zip -NY_BLOCKS_URL=$(NY_ROOT)/tl_2009_36_tabblock00.zip # Starting with PostGIS 1.4.0, these paths are calculated at install # time using the pg_config utility. Rather than try to guess where @@ -28,79 +13,35 @@ PG_SHAREDIR=`pg_config --sharedir` # Necessary to run test/data without prerequisites. # -.PHONY : test data +.PHONY : test data lib # The default task, since it comes first in the list. # -all: clean test +all: clean lib test test: ./bin/run_tests +# Download or check out any third-party libraries. +lib: + if [ ! -d lib/Shapely ]; then \ + wget -O shapely.tar.gz $(SHAPELY_URL); \ + tar -xvzf shapely.tar.gz -C lib/ ; \ + rm shapely.tar.gz; \ + mv lib/Shapely* lib/Shapely; \ + fi; + + # Remove byte-compiled python code. # clean: find ./ -name '*.pyc' -print0 | xargs -0 rm -f -# Download the shapefiles from Tiger if they don't already exist. -# For now, we're only dealing with the Census 2000 Maryland Block -# data, so the filenames are hard-coded. Easy enough to change. -# -data: tiger_blocks tiger_lines - -tiger_blocks: dc_blocks md_blocks va_blocks pa_blocks ny_blocks - -dc_blocks: - mkdir -p data/census2000/dc/block - if [ ! -f data/census2000/dc/block/tl_2009_11_tabblock00.shp ]; \ - then \ - wget -O dcblocks.zip $(DC_BLOCKS_URL); \ - unzip dcblocks.zip -d ./data/census2000/dc/block; \ - rm dcblocks.zip; \ - fi; - -md_blocks: - mkdir -p data/census2000/maryland/block - if [ ! -f data/census2000/maryland/block/tl_2009_24_tabblock00.shp ]; \ - then \ - wget -O mdblocks.zip $(MD_BLOCKS_URL); \ - unzip mdblocks.zip -d ./data/census2000/maryland/block; \ - rm mdblocks.zip; \ - fi; - -va_blocks: - mkdir -p data/census2000/virginia/block - if [ ! -f data/census2000/virginia/block/tl_2009_51_tabblock00.shp ]; \ - then \ - wget -O vablocks.zip $(VA_BLOCKS_URL); \ - unzip vablocks.zip -d ./data/census2000/virginia/block; \ - rm vablocks.zip; \ - fi; - -pa_blocks: - mkdir -p data/census2000/pennsylvania/block - if [ ! -f data/census2000/pennsylvania/block/tl_2009_42_tabblock00.shp ]; \ - then \ - wget -O pablocks.zip $(PA_BLOCKS_URL); \ - unzip pablocks.zip -d ./data/census2000/pennsylvania/block; \ - rm pablocks.zip; \ - fi; - -ny_blocks: - mkdir -p data/census2000/new_york/block - if [ ! -f data/census2000/new_york/block/tl_2009_36_tabblock00.shp ]; \ - then \ - wget -O nyblocks.zip $(NY_BLOCKS_URL); \ - unzip nyblocks.zip -d ./data/census2000/new_york/block; \ - rm nyblocks.zip; \ - fi; - - -tiger_lines: +data: bin/download_data @@ -121,23 +62,27 @@ db: data newdb tiger_blocks_table tiger_lines_table sf1_blocks_table -a \ -s $(TIGER_SRID) \ -D \ - $$state/block/*.shp \ + $$state/blocks/*.shp \ tiger_blocks \ | psql -U $(DB_USER) -d $(DB_NAME); \ done; -# MD Lines +# All Lines # # Since the table and index already exist, we can utilize -a, # and leave -I out. - for x in data/census2000/maryland/lines/*.shp; do \ - $(PG_BINDIR)/shp2pgsql \ - -a \ - -s $(TIGER_SRID) \ - -D \ - $$x \ - tiger_lines \ - | psql -U $(DB_USER) -d $(DB_NAME); \ + for state in data/census2000/*; do \ + for shapefile in $$state/lines/*.shp; do \ + echo "Importing $$shapefile."; \ + $(PG_BINDIR)/shp2pgsql \ + -a \ + -s $(TIGER_SRID) \ + $$shapefile \ + tiger_lines \ + | bin/filter-transactions \ + | psql -U $(DB_USER) -d $(DB_NAME) \ + > /dev/null; \ + done; \ done; bin/sf1blocks2sql src/Tests/Fixtures/SummaryFile1/mdgeo.uf1 sf1_blocks \ @@ -187,7 +132,7 @@ tiger_blocks_table: -p \ -I \ -s $(TIGER_SRID) \ - data/census2000/maryland/block/tl_2009_24_tabblock00.shp \ + data/census2000/maryland/blocks/tl_2009_24_tabblock00.shp \ tiger_blocks \ | psql -U postgres -d $(DB_NAME) \ > /dev/null @@ -204,3 +149,8 @@ tiger_lines_table: tiger_lines \ | psql -U postgres -d $(DB_NAME) \ > /dev/null + +# Add a unique index on the "tlid" column. + psql -U postgres \ + -d census \ + -f sql/create_tlid_unique_index.sql