# Valid book options are a4paper, a5paper, b5paper, letterpaper,
# legalpaper, landscape, 11pt, 12pt, oneside, twocolumn, notitlepage,
# titlepage, openany, draft, fleqn, leqno
+# -P latex.class.options=a5paper
DBLATEX = dblatex \
- -P latex.class.options=a5paper \
- -T db2latex \
- --backend=xetex \
+ -T simple \
+ --indexstyle=myindexstyle.ist \
--param lingua=nb \
--xsl-user=data/user_param.xsl \
--xsl-user=data/xetex_param.xsl \
-p data/pdf.xsl
+
DBTOEPUB = dbtoepub
IMAGES = images/cc.png
-XSLTS = \
- data/user_param.xsl \
- data/xetex_param.xsl \
- data/pdf.xsl
+XSLT = \
+ data/user_param.xsl
+
+DB_XSLT = \
+ data/xetex_param.xsl
+
+HTML_XSLT = \
+ $(XSLT) \
+ data/html.xsl \
+ data/stylesheet-html.xsl
+
+PDF_XSLT = \
+ $(XSLT) \
+ data/lulu.xsl \
+ data/licentia.xsl \
+ data/pdf.xsl \
+ data/stylesheet-fo.xsl
+
all: lint lint.nb html epub pdf
epub: freeculture.nb.epub
html: freeculture.html freeculture.nb.html
-%.pdf: %.xml $(IMAGES) $(XSLTS)
- $(DBLATEX) $<
-
-# Alternative processing path to dblatex is to use xmlto using fop to
-# create PDF like this. The PDF output (visual design) is better, but
-# the PDF index and footnote handling is worse and images are missing.
-# xmlto --noautosize --stringparam paper.type=A5 \
-# --with-fop pdf $<
-
-# Third alternative is to use xsltproc and fop directly, as
-# recommended by <URL: http://www.sagehill.net/docbookxsl/index.html > .
-# This include images, but the index and footnote handling is
-# broken.
-# xsltproc \
-# --output myfile.fo \
-# --stringparam paper.type A5 \
-# /usr/share/xml/docbook/stylesheet/docbook-xsl/fo/docbook.xsl \
-# $<
-# fop -fo myfile.fo -pdf $@
-
-%.html: %.xml $(IMAGES)
- xmlto html-nochunks $<
+%.pdf: %.xml $(IMAGES) $(PDF_XSLT) Makefile
+# Possible pipelines:
+#
+# dblatex:
+# This converts the docbook content to latex and leave it to latex
+# to format it.
+#
+# xmlto:
+# Alternative processing path to dblatex is to use xmlto using fop
+# to create PDF like this. The PDF output (visual design) is
+# better, but the footnote handling is worse and images are missing.
+#
+# docbook-xsl:
+# Third alternative is to use xsltproc and fop directly, as
+# recommended by <URL: http://www.sagehill.net/docbookxsl/index.html >.
+# This include images, but the index refs and footnote handling
+# is broken.
+
+ pipeline=dblatex; \
+ echo "Using $$pipeline pipeline" ; \
+ case "$$pipeline" in \
+ dblatex) \
+ $(DBLATEX) $< ; \
+ ;; \
+ xmlto) \
+ xmlto --noautosize \
+ -x data/stylesheet-fo.xsl \
+ --with-fop pdf $< ; \
+ ;; \
+ docbook-xsl) \
+ xsltproc \
+ --output $(subst .pdf,.fo,$@) \
+ data/stylesheet-fo.xsl \
+ $< ; \
+ fop -c data/fop-params.xconf -fo $(subst .pdf,.fo,$@) -pdf $@ ; \
+ ;; \
+ esac
+
+pdf-compare: freeculture.xml $(IMAGES)
+ dblatex -o freeculture-dblatex.pdf freeculture.xml
+ dblatex -T db2latex \-o freeculture-dblatex-db2latex.pdf freeculture.xml
+
+ # plain xmlto fail
+ #xmlto pdf freeculture.xml && mv freeculture.pdf freeculture-xmlto.pdf
+ xmlto --with-fop pdf freeculture.xml && mv freeculture.pdf freeculture-xmlto-fop.pdf
+
+ xsltproc --output freeculture-docbook-xsl.fo \
+ /usr/share/xml/docbook/stylesheet/docbook-xsl/fo/docbook.xsl \
+ freeculture.xml
+ fop -fo freeculture-docbook-xsl.fo -pdf freeculture-docbook-xsl.pdf
+
+%.html: %.xml $(IMAGES) $(HTML_XSLT)
+ xmlto \
+ -x data/stylesheet-html.xsl \
+ html-nochunks \
+ $<
%.txt: %.xml $(IMAGES)
xmlto txt $<
freeculture.xml:
GET $(url) | gunzip > freeculture.xml
+# <beginpage> workaround can be removed when BTS #684137 is fixed in
+# po4a.
freeculture.pot: freeculture.xml
- po4a-gettextize -f docbook -m $^ > $@.new && mv $@.new $@
+ po4a-gettextize -o nodefault='<beginpage>' -o inline='<beginpage>' \
+ -f docbook -m $^ > $@.new && mv $@.new $@
stats: update-stats progress.png
-update-stats:
+update-stats: freeculture.nb.po lint.nb
( \
printf "%s " $$(date +"%Y-%m-%dT%H%M") ; \
msgfmt -o /dev/null --statistics freeculture.nb.po 2>&1 \