add Makefile with quarto and rdf snippets

2 files changed, 93 insertions, 0 deletions

diff --git a/_make/quarto.mk b/_make/quarto.mk
new file mode 100644
index 0000000..63ad617
--- /dev/null
+++ b/_make/quarto.mk
@@ -0,0 +1,56 @@
+# Make snippet for rendering Quarto documents
+#
+# Copyright 2024, Jonas Smedegaard <dr@jones.dk>
+# SPDX-License-Identifier: GPL-3+
+#
+# Setup:
+# In main Makefile...
+#  * set variable DOCUMENTS
+#  * set variable DOCUMENT_APPENDIX_REGEX if needed
+#  * include this make snippet
+#
+# Dependencies:
+#  * quarto <https://quarto.org/>
+#  * bat <https://github.com/sharkdp/bat>, with executable named "batcat"
+#  * perl v5.10.1 or newer
+
+# list of relative paths to directories,
+# each containing an index.qmd file.
+#DOCUMENTS = \
+# main_paper \
+# promo_article \
+# research/deep/superconductors
+# research/deep/fringe
+
+# regular expressions anchored at the beginning of a single line,
+# to match the beginning of content in document to omit from counting
+DOCUMENT_APPENDIX_REGEX ?= Appendix\\b
+DOCUMENT_BIBLIOGRAPHY_REGEX ?= Bibliography\\b
+
+$(DOCUMENTS:%=doc-render-%): doc-render-%: %.qmd
+	quarto render $<
+
+$(DOCUMENTS:%=doc-screening-of-%): doc-screening-of-%: %.qmd
+	QUARTO_LOG_LEVEL=quiet \
+	quarto render $< --to markdown --output - | batcat --file-name index.qmd --language markdown
+
+# count all characters except horisontal rulers,
+# until appendices
+$(DOCUMENTS:%=doc-charcount-of-%): doc-charcount-of-%: %.qmd
+	QUARTO_LOG_LEVEL=quiet \
+	quarto render $< --to plain --columns=9999 --output - \
+	| perl -nsE 'next if /^-*$$/; $$done += /^$$re/; $$chars += length unless $$done; END { say $$chars }' \
+	 -- -re="$(DOCUMENT_APPENDIX_REGEX)"
+
+# count all characters except horisontal rulers,
+# until bibliography,
+# divide by 2400 and floor the result
+$(DOCUMENTS:%=doc-pagecount-of-%): doc-pagecount-of-%: %.qmd
+	QUARTO_LOG_LEVEL=quiet \
+	quarto render $< --to plain --columns=9999 --output - \
+	| perl -nsE 'next if /^-*$$/; $$done += /^$$re/; $$chars += length unless $$done; END { say $$chars }' \
+	 -- -re="$(DOCUMENT_BIBLIOGRAPHY_REGEX)" \
+	| perl -nE 'say int($$_ / 2400)'
+
+.PHONY: $(DOCUMENTS:%=doc-render-%) $(DOCUMENTS:%=doc-screening-of-%) \
+ $(DOCUMENTS:%=doc-charcount-of-%) $(DOCUMENTS:%=doc-pagecount-of-%)
diff --git a/_make/rdf.mk b/_make/rdf.mk
new file mode 100644
index 0000000..0586ade
--- /dev/null
+++ b/_make/rdf.mk
@@ -0,0 +1,37 @@
+# Make snippet for extracting RDF data from PDF documents
+#
+# Copyright 2024, Jonas Smedegaard <dr@jones.dk>
+# SPDX-License-Identifier: GPL-3+
+#
+# Setup:
+# In main Makefile...
+#  * set variable PDF_DOCUMENTS or use simple and slow default
+#  * set variable BASE_IRI or use default file URI
+#  * include this make snippet
+#
+# Dependencies:
+#  * podofoxmp (e.g. Debian package libpodofo-utils)
+#  * rapper (e.g. Debian package raptor2-utils)
+#  * perl v5.36.0 or newer
+
+# list of relative paths to PDF documents
+#PDF_DOCUMENTS = \
+# main_paper.pdf \
+# promo_article.pdf \
+# research/deep/superconductors/report.pdf
+# research/deep/fringe/index.pdf
+
+PDF_DOCUMENTS ?= $(wildcard *.pdf)
+
+# extract XMP metadata from PDF document,
+# and convert to the human-friendlier RDF/Turtle serialization
+$(PDF_DOCUMENTS:%=turtle-from-%): turtle-from-%:
+	@pdfinfo -meta $* \
+	| perl -gp \
+	 -e 's,\s*<\?xpacket [^>]+>\s*,,;' \
+	 -e 's,<x:xmpmeta [^>]+>,<?xml version="1.0" encoding="utf-8"?>,;' \
+	 -e 's,\s*</x:xmpmeta>,,;' \
+	 -e 's,\s*<\?xpacket [^>]+>,,;' \
+	| rapper -q -i rdfxml -o turtle - $(BASE_IRI)$(dir $*)
+
+.PHONY: $(PDF_DOCUMENTS:%=turtle-from-%)