# This file is in the public domain. # All: build HTML pages in all languages and compile the # TypeScript logic in web-common. # Hardly anyone seems to read README files anymore, so keep this note here: # Don't remove the variables for python etc. They exist # because one system sticks with PEPs, and others opt # for installing every version side-by-side, # Same goes for babel. include config.mk all: css locale template cp -R dist rendered/ cp -R static rendered/ cp rendered/static/robots.txt rendered/robots.txt cp rendered/static/stage.robots.txt rendered/stage.robots.txt cp rendered/static/robots.txt rendered/dist/robots.txt cp rendered/static/robots.txt rendered/en/robots.txt cp rendered/static/robots.txt rendered/de/robots.txt cp rendered/static/robots.txt rendered/es/robots.txt cp rendered/static/robots.txt rendered/fr/robots.txt cp rendered/static/robots.txt rendered/it/robots.txt /bin/sh make_sitemap.sh cp rendered/sitemap.xml rendered/en/sitemap.xml cp rss.xml rendered/rss.xml cp rss.xml rendered/en/rss.xml cp rss.xml rendered/de/rss.xml cp rss.xml rendered/es/rss.xml cp rss.xml rendered/fr/rss.xml cp rss.xml rendered/it/rss.xml cp static/moved.html rendered/frontpage.html cd rendered; ln -fs frontpage.html frontpage cp static/moved_gsoc.html rendered/gsoc.html cd rendered; ln -fs gsoc.html gsoc mkdir rendered/node ; cp static/moved_about.html rendered/node/about.html cd rendered/node ; ln -fs about.html 397 cp static/moved_about.html rendered/about.html cd rendered ; ln -fs about.html philosophy # Extract translateable strings from jinja2 templates. # Because of the local i18nfix extractor module we need # to set the pythonpath before invoking pybabel. locale/messages.pot: common/*.j2.inc template/*.j2 PYTHONPATH=. $(BABEL) -v extract -F locale/babel.map -o locale/messages.pot . # Update translation (.po) files with new strings. locale-update: locale/messages.pot msgmerge -U -m --previous locale/en/LC_MESSAGES/messages.po locale/messages.pot msgmerge -U -m --previous locale/de/LC_MESSAGES/messages.po locale/messages.pot msgmerge -U -m --previous locale/fr/LC_MESSAGES/messages.po locale/messages.pot msgmerge -U -m --previous locale/es/LC_MESSAGES/messages.po locale/messages.pot msgmerge -U -m --previous locale/it/LC_MESSAGES/messages.po locale/messages.pot if grep -nA1 '#-#-#-#-#' locale/*/LC_MESSAGES/messages.po; then echo -e "\nERROR: Conflicts encountered in PO files.\n"; exit 1; fi # sass preprocessor css: sassc static/styles.sass static/styles.css # Compile translation files for use. locale-compile: $(BABEL) -v compile -d locale -l en --use-fuzzy $(BABEL) -v compile -d locale -l de --use-fuzzy $(BABEL) -v compile -d locale -l fr --use-fuzzy $(BABEL) -v compile -d locale -l it --use-fuzzy $(BABEL) -v compile -d locale -l es --use-fuzzy # Process everything related to gettext translations. locale: locale-update locale-compile # Run the jinja2 templating engine to expand templates to HTML # incorporating translations. template: locale-compile $(PYTHON) ./template.py it: template current_dir = $(shell pwd) run: all @[ "$(BROWSER)" ] || ( echo "You need to export the environment variable 'BROWSER' to run this."; exit 1 ) $(RUN_BROWSER) http://0.0.0.0:8000 & cd rendered && $(PYTHON) -m http.server # docker-all: Build using a docker image which contains all the needed packages. docker: docker-all docker-all: docker build -t gnunet-www-builder . # Importing via the shell like this is hacky, # but after trying lots of other ways, this works most reliably... $(PYTHON) -c 'import i18nfix' docker run --rm -v $$(pwd):/tmp/ --user $$(id -u):$$(id -g) gnunet-www-builder clean: rm -rf __pycache__ rm -rf en/ de/ fr/ it/ es/ ru/ rm -rf rendered/ rm -rf *.pyc *~ \.*~ \#*\#