X-Git-Url: http://gitweb.michael.orlitzky.com/?a=blobdiff_plain;f=makefile;h=d32c27f81ef2467a6b513e4fbb4876b494c3a101;hb=11f2b70a38c7a8dbed276cd420a94e0e24bd3084;hp=36543d4e156bb7dd5a6ea8c60e948d613dd29739;hpb=d6aa9f176f67ee918f854c1df3ec5b072a502288;p=dead%2Fcensus-tools.git diff --git a/makefile b/makefile index 36543d4..d32c27f 100644 --- a/makefile +++ b/makefile @@ -1,7 +1,56 @@ -DB_NAME='census' -DB_USER='postgres' -TIGER_DATA_URL='http://www2.census.gov/geo/tiger/TIGER2009/24_MARYLAND/tl_2009_24_tabblock00.zip' -TIGER_SRID='4269' +DB_NAME=census +DB_USER=postgres +TIGER_SRID=4269 + +# 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 + +# URLs for the DC county all-lines shapefiles. +# D.C. just has one Census county. +DC_LINES_URL=$(DC_ROOT)/11001_District_of_Columbia/tl_2009_11001_edges.zip + +# This is where it gets ugly. +# +# URLs for the MD county all-lines shapefiles. +ALLEGANY_LINES_URL=$(MD_ROOT)/24001_Allegany_County/tl_2009_24001_edges.zip +ANNE_ARUNDEL_LINES_URL=$(MD_ROOT)/24003_Anne_Arundel_County/tl_2009_24003_edges.zip +BALTIMORE_CO_LINES_URL=$(MD_ROOT)/24005_Baltimore_County/tl_2009_24005_edges.zip +BALTIMORE_CI_LINES_URL=$(MD_ROOT)/24510_Baltimore_city/tl_2009_24510_edges.zip +CALVERT_LINES_URL=$(MD_ROOT)/24009_Calvert_County/tl_2009_24009_edges.zip +CAROLINE_LINES_URL=$(MD_ROOT)/24011_Caroline_County/tl_2009_24011_edges.zip +CARROLL_LINES_URL=$(MD_ROOT)/24013_Carroll_County/tl_2009_24013_edges.zip +CECIL_LINES_URL=$(MD_ROOT)/24015_Cecil_County/tl_2009_24015_edges.zip +CHARLES_LINES_URL=$(MD_ROOT)/24017_Charles_County/tl_2009_24017_edges.zip +DORCHESTER_LINES_URL=$(MD_ROOT)/24019_Dorchester_County/tl_2009_24019_edges.zip +FREDERICK_LINES_URL=$(MD_ROOT)/24021_Frederick_County/tl_2009_24021_edges.zip +GARRETT_LINES_URL=$(MD_ROOT)/24023_Garrett_County/tl_2009_24023_edges.zip +HARFORD_LINES_URL=$(MD_ROOT)/24025_Harford_County/tl_2009_24025_edges.zip +HOWARD_LINES_URL=$(MD_ROOT)/24027_Howard_County/tl_2009_24027_edges.zip +KENT_LINES_URL=$(MD_ROOT)/24029_Kent_County/tl_2009_24029_edges.zip +MONTGOMERY_LINES_URL=$(MD_ROOT)/24031_Montgomery_County/tl_2009_24031_edges.zip +PRINCE_GEORGES_LINES_URL=$(MD_ROOT)/24033_Prince_Georges_County/tl_2009_24033_edges.zip +QUEEN_ANNES_LINES_URL=$(MD_ROOT)/24035_Queen_Annes_County/tl_2009_24035_edges.zip +ST_MARYS_LINES_URL=$(MD_ROOT)/24037_St_Marys_County/tl_2009_24037_edges.zip +SOMERSET_LINES_URL=$(MD_ROOT)/24039_Somerset_County/tl_2009_24039_edges.zip +TALBOT_LINES_URL=$(MD_ROOT)/24041_Talbot_County/tl_2009_24041_edges.zip +WASHINGTON_LINES_URL=$(MD_ROOT)/24043_Washington_County/tl_2009_24043_edges.zip +WICOMICO_LINES_URL=$(MD_ROOT)/24045_Wicomico_County/tl_2009_24045_edges.zip +WORCESTER_LINES_URL=$(MD_ROOT)/24047_Worcester_County/tl_2009_24047_edges.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 @@ -34,56 +83,383 @@ clean: # For now, we're only dealing with the Census 2000 Maryland Block # data, so the filenames are hard-coded. Easy enough to change. # -data: +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 tmp.zip $(TIGER_DATA_URL); \ - unzip tmp.zip -d ./data/census2000/maryland/block; \ - rm tmp.zip; \ + 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: dc_lines md_lines + +dc_lines: + mkdir -p data/census2000/dc/lines + if [ ! -f data/census2000/dc/lines/tl_2009_11001_edges.shp ]; \ + then \ + wget -O dclines.zip $(DC_LINES_URL); \ + unzip dclines.zip -d ./data/census2000/dc/lines; \ + rm dclines.zip; \ + fi; + +md_lines: allegany_lines anne_arundel_lines baltimore_co_lines baltimore_ci_lines calvert_lines caroline_lines carroll_lines cecil_lines charles_lines dorchester_lines frederick_lines garrett_lines harford_lines howard_lines kent_lines montgomery_lines prince_georges_lines queen_annes_lines st_marys_lines somerset_lines talbot_lines washington_lines wicomico_lines worcester_lines + +allegany_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24001_edges.shp ]; \ + then \ + wget -O alleganylines.zip $(ALLEGANY_LINES_URL); \ + unzip alleganylines.zip -d ./data/census2000/maryland/lines; \ + rm alleganylines.zip; \ + fi; + +anne_arundel_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24003_edges.shp ]; \ + then \ + wget -O aalines.zip $(ANNE_ARUNDEL_LINES_URL); \ + unzip aalines.zip -d ./data/census2000/maryland/lines; \ + rm aalines.zip; \ + fi; + +baltimore_co_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24005_edges.shp ]; \ + then \ + wget -O bcolines.zip $(BALTIMORE_CO_LINES_URL); \ + unzip bcolines.zip -d ./data/census2000/maryland/lines; \ + rm bcolines.zip; \ + fi; + +baltimore_ci_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24510_edges.shp ]; \ + then \ + wget -O bcilines.zip $(BALTIMORE_CI_LINES_URL); \ + unzip bcilines.zip -d ./data/census2000/maryland/lines; \ + rm bcilines.zip; \ + fi; + +calvert_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24009_edges.shp ]; \ + then \ + wget -O calvertlines.zip $(CALVERT_LINES_URL); \ + unzip calvertlines.zip -d ./data/census2000/maryland/lines; \ + rm calvertlines.zip; \ + fi; + +caroline_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24011_edges.shp ]; \ + then \ + wget -O carolinelines.zip $(CAROLINE_LINES_URL); \ + unzip carolinelines.zip -d ./data/census2000/maryland/lines; \ + rm carolinelines.zip; \ + fi; + +carroll_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24013_edges.shp ]; \ + then \ + wget -O carrolllines.zip $(CARROLL_LINES_URL); \ + unzip carrolllines.zip -d ./data/census2000/maryland/lines; \ + rm carrolllines.zip; \ + fi; + +cecil_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24015_edges.shp ]; \ + then \ + wget -O cecillines.zip $(CECIL_LINES_URL); \ + unzip cecillines.zip -d ./data/census2000/maryland/lines; \ + rm cecillines.zip; \ + fi; + +charles_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24017_edges.shp ]; \ + then \ + wget -O charleslines.zip $(CHARLES_LINES_URL); \ + unzip charleslines.zip -d ./data/census2000/maryland/lines; \ + rm charleslines.zip; \ + fi; + +dorchester_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24019_edges.shp ]; \ + then \ + wget -O dorchesterlines.zip $(DORCHESTER_LINES_URL); \ + unzip dorchesterlines.zip -d ./data/census2000/maryland/lines; \ + rm dorchesterlines.zip; \ + fi; + +frederick_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24021_edges.shp ]; \ + then \ + wget -O fredericklines.zip $(FREDERICK_LINES_URL); \ + unzip fredericklines.zip -d ./data/census2000/maryland/lines; \ + rm fredericklines.zip; \ + fi; + +garrett_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24023_edges.shp ]; \ + then \ + wget -O garrettlines.zip $(GARRETT_LINES_URL); \ + unzip garrettlines.zip -d ./data/census2000/maryland/lines; \ + rm garrettlines.zip; \ + fi; -# This task does a couple of things. First, it drops and re-creates -# the DB_NAME database (or schema, whatever). Then, it adds PL/pgSQL -# support to the database. +harford_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24025_edges.shp ]; \ + then \ + wget -O harfordlines.zip $(HARFORD_LINES_URL); \ + unzip harfordlines.zip -d ./data/census2000/maryland/lines; \ + rm harfordlines.zip; \ + fi; + +howard_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24027_edges.shp ]; \ + then \ + wget -O howardlines.zip $(HOWARD_LINES_URL); \ + unzip howardlines.zip -d ./data/census2000/maryland/lines; \ + rm howardlines.zip; \ + fi; + +kent_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24029_edges.shp ]; \ + then \ + wget -O kentlines.zip $(KENT_LINES_URL); \ + unzip kentlines.zip -d ./data/census2000/maryland/lines; \ + rm kentlines.zip; \ + fi; + +montgomery_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24031_edges.shp ]; \ + then \ + wget -O montgomerylines.zip $(MONTGOMERY_LINES_URL); \ + unzip montgomerylines.zip -d ./data/census2000/maryland/lines; \ + rm montgomerylines.zip; \ + fi; + +prince_georges_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24033_edges.shp ]; \ + then \ + wget -O pglines.zip $(PRINCE_GEORGES_LINES_URL); \ + unzip pglines.zip -d ./data/census2000/maryland/lines; \ + rm pglines.zip; \ + fi; + +queen_annes_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24035_edges.shp ]; \ + then \ + wget -O qalines.zip $(QUEEN_ANNES_LINES_URL); \ + unzip qalines.zip -d ./data/census2000/maryland/lines; \ + rm qalines.zip; \ + fi; + +st_marys_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24037_edges.shp ]; \ + then \ + wget -O smlines.zip $(ST_MARYS_LINES_URL); \ + unzip smlines.zip -d ./data/census2000/maryland/lines; \ + rm smlines.zip; \ + fi; + +somerset_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24039_edges.shp ]; \ + then \ + wget -O somersetlines.zip $(SOMERSET_LINES_URL); \ + unzip somersetlines.zip -d ./data/census2000/maryland/lines; \ + rm somersetlines.zip; \ + fi; + +talbot_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24041_edges.shp ]; \ + then \ + wget -O talbotlines.zip $(TALBOT_LINES_URL); \ + unzip talbotlines.zip -d ./data/census2000/maryland/lines; \ + rm talbotlines.zip; \ + fi; + +washington_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24043_edges.shp ]; \ + then \ + wget -O washingtonlines.zip $(WASHINGTON_LINES_URL); \ + unzip washingtonlines.zip -d ./data/census2000/maryland/lines; \ + rm washingtonlines.zip; \ + fi; + +wicomico_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24045_edges.shp ]; \ + then \ + wget -O wicomicolines.zip $(WICOMICO_LINES_URL); \ + unzip wicomicolines.zip -d ./data/census2000/maryland/lines; \ + rm wicomicolines.zip; \ + fi; + +worcester_lines: + mkdir -p data/census2000/maryland/lines + if [ ! -f data/census2000/maryland/lines/tl_2009_24047_edges.shp ]; \ + then \ + wget -O worcesterlines.zip $(WORCESTER_LINES_URL); \ + unzip worcesterlines.zip -d ./data/census2000/maryland/lines; \ + rm worcesterlines.zip; \ + fi; + + +# This imports the Tiger data using shp2pgsql. The shapefiles +# should exist, since this task depends on the "data" task, which +# downloads said shapefiles. +# +# After the TIGER import is done, we use the sf1blocks2sql script to +# parse and import the geographic header record information. +# +db: data newdb tiger_blocks_table tiger_lines_table sf1_blocks_table +# All Blocks +# +# The table already exists, so we can append to it, and we don't have +# to create the GiST index. + for state in data/census2000/*; do \ + $(PG_BINDIR)/shp2pgsql \ + -a \ + -s $(TIGER_SRID) \ + -D \ + $$state/block/*.shp \ + tiger_blocks \ + | psql -U $(DB_USER) -d $(DB_NAME); \ + done; + +# MD Lines +# +# Since the table and index already exist, we can utilize -a, +# and leave -I out. + for x in data/census2000/maryland/lines/*.shp; do \ + $(PG_BINDIR)/shp2pgsql \ + -a \ + -s $(TIGER_SRID) \ + -D \ + $$x \ + tiger_lines \ + | psql -U $(DB_USER) -d $(DB_NAME); \ + done; + + bin/sf1blocks2sql src/Tests/Fixtures/SummaryFile1/mdgeo.uf1 sf1_blocks \ + | psql -U postgres -d $(DB_NAME) \ + > /dev/null + + + +# First, we drop and re-create the DB_NAME database (or schema, +# whatever). Then, we add PL/pgSQL support to the database. # # At that point, we import the two PostGIS files, postgis.sql and # spatial_ref_sys.sql. The postgis.sql file contains the geometry # functions, while spatial_ref_sys.sql contains a table of SRIDs, and # their associated properties. PostGIS requires both. # -# Then, we import the Tiger data using shp2pgsql. The shapefiles -# should exist, since this task depends on the "data" task, which -# downloads said shapefiles. -# -# Finally, we create the table for the demographic data (obtained from -# the geographic header records), and populate that table with the output -# of the sf1blocks2sql script. -# -db: data - # Ignore the result of dropdb when it fails. +newdb: +# Ignore the result of dropdb when it fails. dropdb -U $(DB_USER) $(DB_NAME) || true createdb -U $(DB_USER) $(DB_NAME) createlang -U $(DB_USER) plpgsql $(DB_NAME) - psql -d $(DB_NAME) \ - -U $(DB_USER) \ - -f $(PG_SHAREDIR)/contrib/postgis.sql + psql -d $(DB_NAME) \ + -U $(DB_USER) \ + -f $(PG_SHAREDIR)/contrib/postgis.sql \ + > /dev/null - psql -d $(DB_NAME) \ - -U $(DB_USER) \ - -f $(PG_SHAREDIR)/contrib/spatial_ref_sys.sql + psql -d $(DB_NAME) \ + -U $(DB_USER) \ + -f $(PG_SHAREDIR)/contrib/spatial_ref_sys.sql \ + > /dev/null - $(PG_BINDIR)/shp2pgsql -I \ - -s $(TIGER_SRID) \ - data/census2000/maryland/block/tl_2009_24_tabblock00.shp \ - tiger \ - | psql -U $(DB_USER) -d $(DB_NAME) - psql -d $(DB_NAME) \ - -U $(DB_USER) \ - -f sql/create-sf1_blocks-table.sql +# This just runs the SQL script to create the sf1_blocks table. +sf1_blocks_table: + psql -d $(DB_NAME) \ + -U $(DB_USER) \ + -f sql/create-sf1_blocks-table.sql \ + > /dev/null + + +# Create the tiger_blocks table, and create its GiST index. Having the +# table already exist makes importing via shp2pgsql much easier. +# Any blocks file will work as an argument. +tiger_blocks_table: + $(PG_BINDIR)/shp2pgsql \ + -p \ + -I \ + -s $(TIGER_SRID) \ + data/census2000/maryland/block/tl_2009_24_tabblock00.shp \ + tiger_blocks \ + | psql -U postgres -d $(DB_NAME) \ + > /dev/null - bin/sf1blocks2sql src/Tests/Fixtures/SummaryFile1/mdgeo.uf1 sf1_blocks \ - | psql -U postgres -d $(DB_NAME) +# Prepare the tiger_lines table, and create the GiST index on its +# geometry column. Any lines shapefile will do here. +tiger_lines_table: + $(PG_BINDIR)/shp2pgsql \ + -p \ + -I \ + -s $(TIGER_SRID) \ + data/census2000/maryland/lines/tl_2009_24510_edges.shp \ + tiger_lines \ + | psql -U postgres -d $(DB_NAME) \ + > /dev/null