docs: hack a charset=utf-8 into pwg/adm html versions

So the encoding of the original document is respected and
displays properly in browsers where the encoding autodetection
fails to recognise that it's utf-8.

https://bugzilla.gnome.org/show_bug.cgi?id=639448
This commit is contained in:
Tim-Philipp Müller 2011-01-14 09:21:23 +00:00
parent 9382017b42
commit 0177ff171a

View file

@ -89,6 +89,7 @@ $(BUILDDIR)/$(MAIN): $(XML) $(CSS) $(EXTRA_SRC)
# we should switch to xsltproc
# docbook2html aka jade can't add the encoding easily to the html meta
# (but we are lazy and just abuse sed to add it)
# jw -f docbook -b html -d pwg.dsl -o ../html -V '%use-id-as-filename%' $(MAIN)
# this is a starting point
# xsltproc --nonet /usr/share/xml/docbook/stylesheet/nwalsh/html/docbook.xsl pwg.xml
@ -98,7 +99,8 @@ html/index.html: $(BUILDDIR)/$(MAIN) $(PNG_BUILT) $(FIG_SRC)
@echo "*** Generating HTML output ***"
@-mkdir -p html
@cp -f $(srcdir)/../image-png $(BUILDDIR)/image.entities
@cd $(BUILDDIR) && docbook2html -o ../html -V '%use-id-as-filename%' $(MAIN)
@cd $(BUILDDIR) && SP_ENCODING="UTF-8" docbook2html -o ../html -V '%use-id-as-filename%' $(MAIN)
@$(SED) -i -e 's/\(^CONTENT.*\)\(.>\)/\1;charset=UTF-8\2/' html/*html
@test "x$(CSS)" != "x" && \
echo "Copying .css files: $(CSS)" && \
cp $(srcdir)/$(CSS) html