Last active
April 17, 2020 16:08
-
-
Save danielbraun/e1534ce5da56894c4fce15a16016059b to your computer and use it in GitHub Desktop.
My magic makefile to make everything happen
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
SPACY_MODEL ?= xx_ent_wiki_sm | |
PYTHONPATH ?= ~/django_home | |
DJANGO_SETTINGS_MODULE ?= django_home.settings | |
%.min.js: %.js | |
uglifyjs < $< > $@ | |
%.xml: %.edn | |
edn2xml < $< > $@ | |
%.bundle.js: %.js | |
browserify $< > $@ | |
%.tab: %.mysql.sql | |
mysql --defaults-file=my.cnf < $< > $@ | |
%.xml: %.mysql.sql | |
mysql --defaults-file=my.cnf --xml < $< > $@ | |
%.html: %.psql.sql | |
psql --html < $< > $@ | |
%.xml: %.psql.sql | |
# TODO fix double quotes | |
echo "select concat('<?xml version=\"1.0\"?>',query_to_xml(:q, false, false, ''));" \ | |
| psql -t -A -v q="'$(shell cat $<)'" \ | |
> $@ | |
%.xml: %.html | |
xsltproc --html $*.xsl $< -o $@ | |
%.txt: %.pdf | |
pdftotext - - < $< > $@ | |
%.info.txt: %.pdf | |
pdfinfo $< > $@ | |
%.pdf: %.svg | |
cairosvg $< -o $@ | |
%.out: %.curl | |
curl -K $< > $@ | |
%.models.py: | |
PYTHONPATH=$(PYTHONPATH) DJANGO_SETTINGS_MODULE=$(DJANGO_SETTINGS_MODULE) python3 -m django inspectdb $(shell basename $*) > $@ | |
U.S.DividendChampions.xlsx: | |
curl -L https://bitly.com/USDividendChampions > $@ | |
%.dividends.csv: | |
curl "https://query1.finance.yahoo.com/v7/finance/download/$(shell basename $*)?period1=0&period2=$(shell date +%s)&interval=1d&events=div" > $@ | |
%.sorted.csv: %.csv | |
csvsort < $< > $@ | |
%/stocks.nyse.min.json: | |
curl 'https://www.nyse.com/api/quotes/filter' \ | |
-XPOST \ | |
-H 'Content-Type: application/json' \ | |
--data-binary '{"instrumentType":"EQUITY","pageNumber":1,"maxResultsPerPage":100000}' > $@ | |
%/dividends: %/stock_symbols.nyse.csv.tail | |
mkdir -p $@ | |
cat $< | xargs printf "$@/%s.dividends.sorted.csv\n" | xargs $(MAKE) | |
%/stock_symbols.nyse.csv: %/stocks.nyse.csv | |
cat $< | csvcut -c symbolTicker > $@ | |
%.tail: % | |
tail -n +2 < $< > $@ | |
%.csv: %.json | |
json2csv < $< > $@ | |
%.json: %.iob | |
python3 -m spacy convert $< > $@ | |
%.table.txt: %.csv | |
csvlook $@ > $@ | |
%.json: %.min.json | |
jq < $< > $@ | |
%.min.json: %.csv | |
csvjson < $< > $@ | |
%.out.edn: %.edn | |
%.xml: %.php | |
php < $< > $@ | |
%.svg: %.csv | |
gnuplot -e "set key autotitle columnhead; set datafile separator ','; set terminal svg enhanced mouse jsdir 'http://gnuplot.sourceforge.net/demo_svg_5.2'; plot '$<';" > $@ | |
%.instagram.jpg: | |
curl -L "https://www.instagram.com/p/$(shell basename $*)/media?size=l" > $@ | |
# Sign in doesn't work | |
music/%: | |
@mkdir -p `dirname $@` | |
printf "<iframe \ | |
src='https://embed.music.apple.com/us/album/$(shell basename $*)' \ | |
height='100%%' \ | |
frameborder='0' \ | |
allow='autoplay *; encrypted-media *;' \ | |
sandbox='allow-forms allow-popups allow-same-origin allow-scripts allow-storage-access-by-user-activation allow-top-navigation-by-user-activation' \ | |
width='100%%' \ | |
></iframe>" > $@ | |
youtube/%: | |
@mkdir -p `dirname $@` | |
printf "<iframe \ | |
src='https://www.youtube.com/embed/$(shell basename $*)' \ | |
height='100%%' \ | |
frameborder='0' \ | |
width='100%%' \ | |
></iframe>" > $@ | |
yad2/rentals/%.json: | |
@mkdir -p `dirname $@` | |
@res="$(shell curl 'https://www.yad2.co.il/api/pre-load/getFeedIndex/realestate/rent?page=$*' \ | |
-H 'User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_6) AppleWebKit/$*.1.15 (KHTML, like Gecko) Version/12.1.2 Safari/605.1.15' \ | |
-f \ | |
-L \ | |
-s \ | |
-w %{url_effective} \ | |
-o $@)" \ | |
; file --mime $@ | grep "text/html" \ | |
&& echo "Error: Verification required" \ | |
&& rm $@ && open $$res\ | |
&& exit 1 \ | |
|| exit 0 | |
domains/%: | |
mkdir -p $(shell dirname $@) | |
whois $* > $@ | |
nslookup $* >> $@ | |
%.svg: %.geojson | |
geo2svg < $< > $@ | |
%.kml: %.geojson | |
./node_modules/.bin/tokml $< > $@ | |
server: | |
shell2http -cgi / 'export FILE=.`echo $$REQUEST_URI | sed "s/?.*//"` && make -s $$FILE && cat $$FILE' | |
corona.geojson: | |
curl "https://services5.arcgis.com/dlrDjz89gx9qyfev/arcgis/rest/services/Corona_Exposure_View/FeatureServer/0/query?where=1%3D1&objectIds=&time=&geometry=&geometryType=esriGeometryEnvelope&inSR=&spatialRel=esriSpatialRelIntersects&resultType=none&distance=0.0&units=esriSRUnit_Meter&returnGeodetic=false&outFields=*&returnGeometry=true&featureEncoding=esriDefault&multipatchOption=xyFootprint&maxAllowableOffset=&geometryPrecision=&outSR=&datumTransformation=&applyVCSProjection=false&returnIdsOnly=false&returnUniqueIdsOnly=false&returnCountOnly=false&returnExtentOnly=false&returnQueryGeometry=false&returnDistinctValues=false&cacheHint=false&orderByFields=&groupByFieldsForStatistics=&outStatistics=&having=&resultOffset=&resultRecordCount=&returnZ=false&returnM=false&returnExceededLimitFeatures=true&quantizationParameters=&sqlFormat=none&f=pgeojson&token=" -o $@ | |
%.leaflet.json: %.geojson | |
jq -s ".[0].geoJSON=.[1] | first" leaflet.json $< > $@ | |
%.json: %.xml | |
xml2json < $< > $@ | |
%.json.js: %.json | |
browserify --require ./$< -s $(shell basename $*) > $@ | |
%.json.xml: %.json | |
echo "<?xml version=\"1.0\" encoding=\"UTF-8\"?><?xml-stylesheet href=\"leaflet.xsl\" type=\"text/xsl\"?><root><![CDATA[$$(cat $<)]]></root>" > $@ | |
%.leaflet.html: $.leaflet.json | |
%.sql: %.csv | |
csvsql $< > $@ | |
airbnb/%: | |
curl "https://www.airbnb.com/api/v2/pdp_listing_booking_details?_format=for_web_with_date&check_in=2020-03-29&check_out=2020-04-21¤cy=ILS&guests=1&key=d306zoyjsyarp7ifhu67rjxn52tv0t20&listing_id=$*&locale=en&number_of_adults=1&number_of_children=0&number_of_infants=0" \ | |
| jq ".pdp_listing_booking_details[0].price.total.amount" \ | |
> $@ | |
%.spacy.json: %.txt | |
python3 -c'import spacy, sys, json; print(json.dumps(spacy.load("$(SPACY_MODEL)")("".join(sys.stdin.readlines())).to_json()))' < $< | jq > $@ | |
%.ent.spacy.html: %.txt | |
python3 -c 'import spacy, sys; print(spacy.displacy.render([spacy.load("$(SPACY_MODEL)")("".join(sys.stdin.readlines()))], style="ent", page=True))' < $< > $@ | |
%.png: %.pdf | |
cd $(shell dirname $@) && pdftocairo -png -singlefile $(shell basename $<) | |
dividend-aristocrats.csv: | |
curl -L http://www.simplysafedividends.com/intelligent-income/idea_lists/3-dividend-aristocrats.csv > $@ | |
%.test.iob: %.iob | |
cat $< | head -n $(shell echo $(shell cat $< | wc -l) \* 0.6 / 1 | bc) > $@ | |
%.dev.iob: %.iob | |
cat $< | head -n $(shell echo $(shell cat $< | wc -l) \* 0.4 / 1 | bc) > $@ | |
%.model: %.test.json %.dev.json | |
rm -rf $@ | |
python3 -m spacy train \ | |
-p ner \ | |
-b xx_ent_wiki_sm\ | |
xx $@ $*.test.json $*.dev.json | |
clean: | |
git clean -fdX | |
.PHONY: %/dividends *.php clean |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment