More export tools by James I think.
authorRobin H. Johnson <rjohnson@sitka.bclibraries.ca>
Thu, 16 Aug 2012 05:44:02 +0000 (22:44 -0700)
committerRobin H. Johnson <rjohnson@sitka.bclibraries.ca>
Thu, 16 Aug 2012 05:44:02 +0000 (22:44 -0700)
marc_export_custom/sitka-full-export-parallel-bkct.sh [new file with mode: 0755]
marc_export_custom/sitka-full-export-parallel-ms.sh [new file with mode: 0755]

diff --git a/marc_export_custom/sitka-full-export-parallel-bkct.sh b/marc_export_custom/sitka-full-export-parallel-bkct.sh
new file mode 100755 (executable)
index 0000000..1636668
--- /dev/null
@@ -0,0 +1,67 @@
+#!/bin/bash
+
+DIR=`dirname $0`
+#cd /srv/openils/bin
+
+PGUSER=evergreen
+PGHOST=db1.sitka.bclibraries.ca
+PSQL="psql -A -t -U $PGUSER -h $PGHOST"
+DATABASE=evergreen
+#FULL_QUERY="select distinct id from biblio.record_entry where not deleted"
+FULL_QUERY="select distinct acn.record FROM asset.call_number acn join biblio.record_entry bre on (acn.record = bre.id) where not bre.deleted and bre.active and not acn.deleted"
+FULL_QUERY="SELECT DISTINCT acn.record FROM asset.call_number acn JOIN biblio.record_entry bre ON (acn.record = bre.id) WHERE NOT bre.deleted AND bre.active AND not acn.deleted AND owning_lib IN (SELECT id FROM actor.org_unit_descendants(50))"
+POSTSEC_QUERY="select distinct record FROM asset.call_number where owning_lib in (select distinct id from actor.org_unit_descendants(27)) and not deleted and label != '##URI##';"
+NONEXPORTED_QUERY="select distinct acn.record FROM asset.call_number acn join biblio.record_entry bre on (acn.record = bre.id) where bre.deleted or not bre.active or acn.deleted"
+MARC_PARAMS="--config /srv/openils/conf/opensrf_core.xml --items --location SITKA --collapse_to_depth 2 --timeout 300 --force901 --encoding UTF-8 --onlyholdings"
+MARC_EXPORT_BIN="${DIR}/marc_export_custom"
+MARC_EXPORT="${MARC_EXPORT_BIN} ${MARC_PARAMS}"
+
+[ -z "${T}" ] && T=$(date -u +%Y%m%dT%H%M%SZ)
+[ -z "${OUTDIR}" ] && OUTDIR=$(mktemp --tmpdir -d outlook_export.XXXXXXX)
+
+process_batch() {
+       local f="$1" sql="$2" split_count="$3"
+       if [ ! -f ${OUTDIR}/${f}-${T}.id ]; then
+               echo "Getting IDs"
+               $PSQL -c "$sql" $DATABASE | sort | uniq >${OUTDIR}/${f}-${T}.id
+       fi
+       if [ ! -f ${OUTDIR}/${f}.id.targets ]; then
+               echo "Splitting work"
+               split_suffix ${OUTDIR}/${f}-${T}.id ${OUTDIR}/${f}-${T}-split. .id $split_count
+               echo "Making targets"
+               find ${OUTDIR} -name "${f}-${T}-split.[0-9]*[0-9].id" \
+               | sed 's,.id$,.marc,g' \
+               >${OUTDIR}/${f}.id.targets
+       fi
+
+       make ${MAKEOPTS} -f Makefile-marc_export \
+       MARC_EXPORT_BIN="${MARC_EXPORT_BIN}" \
+       MARC_PARAMS="$MARC_PARAMS --exclusion_ini ${f}.ini" \
+       $(<${OUTDIR}/${f}.id.targets)
+
+       find ${OUTDIR} -name "${f}-${T}-split*.marc" \
+       -exec cat \{} \; \
+       >${OUTDIR}/${f}-${T}.marc
+}
+
+split_suffix() {
+       local input="$1" output_prefix="$2" output_suffix="$3" split_count="$4"
+       split -a 7 -d -C ${split_count} ${input} ${output_prefix}
+       find $(dirname ${output_prefix}) -name "$(basename ${output_prefix})*" -exec mv -f \{\} \{\}${output_suffix} \;
+}
+
+# media for BC ELN post-secondary libraries
+export SPLIT_COUNT=10000
+F=bc_eln_media SQL="$POSTSEC_QUERY"
+#process_batch "$F" "${SQL}" $SPLIT_COUNT
+
+# serials for BC ELN post-secondary libraries
+export SPLIT_COUNT=10000
+F=bc_eln_serials SQL="$POSTSEC_QUERY"
+#process_batch "$F" "${SQL}" $SPLIT_COUNT
+
+export SPLIT_COUNT=50000
+# full dump for outlook public libraries
+F=sitka_full_outlook SQL="$FULL_QUERY"
+process_batch "$F" "${SQL}" $SPLIT_COUNT
+
diff --git a/marc_export_custom/sitka-full-export-parallel-ms.sh b/marc_export_custom/sitka-full-export-parallel-ms.sh
new file mode 100755 (executable)
index 0000000..ca4abbb
--- /dev/null
@@ -0,0 +1,66 @@
+#!/bin/bash
+
+DIR=`dirname $0`
+#cd /srv/openils/bin
+
+PGUSER=evergreen
+PGHOST=db1.sitka.bclibraries.ca
+PSQL="psql -A -t -U $PGUSER -h $PGHOST"
+DATABASE=evergreen
+#FULL_QUERY="select distinct id from biblio.record_entry where not deleted"
+FULL_QUERY="select distinct acn.record FROM asset.call_number acn join biblio.record_entry bre on (acn.record = bre.id) where not bre.deleted and bre.active and not acn.deleted"
+POSTSEC_QUERY="select distinct record FROM asset.call_number where owning_lib in (select distinct id from actor.org_unit_descendants(27)) and not deleted and label != '##URI##';"
+NONEXPORTED_QUERY="select distinct acn.record FROM asset.call_number acn join biblio.record_entry bre on (acn.record = bre.id) where bre.deleted or not bre.active or acn.deleted"
+MARC_PARAMS="--config /srv/openils/conf/opensrf_core.xml --items --location SITKA --collapse_to_depth 2 --timeout 300 --force901 --encoding UTF-8 --onlyholdings --cap=12"
+MARC_EXPORT_BIN="${DIR}/marc_export_custom_ms"
+MARC_EXPORT="${MARC_EXPORT_BIN} ${MARC_PARAMS}"
+
+[ -z "${T}" ] && T=$(date -u +%Y%m%dT%H%M%SZ)
+[ -z "${OUTDIR}" ] && OUTDIR=$(mktemp --tmpdir -d outlook_export.XXXXXXX)
+
+process_batch() {
+       local f="$1" sql="$2" split_count="$3"
+       if [ ! -f ${OUTDIR}/${f}-${T}.id ]; then
+               echo "Getting IDs"
+               $PSQL -c "$sql" $DATABASE | sort | uniq >${OUTDIR}/${f}-${T}.id
+       fi
+       if [ ! -f ${OUTDIR}/${f}.id.targets ]; then
+               echo "Splitting work"
+               split_suffix ${OUTDIR}/${f}-${T}.id ${OUTDIR}/${f}-${T}-split. .id $split_count
+               echo "Making targets"
+               find ${OUTDIR} -name "${f}-${T}-split.[0-9]*[0-9].id" \
+               | sed 's,.id$,.marc,g' \
+               >${OUTDIR}/${f}.id.targets
+       fi
+
+       make ${MAKEOPTS} -f Makefile-marc_export \
+       MARC_EXPORT_BIN="${MARC_EXPORT_BIN}" \
+       MARC_PARAMS="$MARC_PARAMS --exclusion_ini ${f}.ini" \
+       $(<${OUTDIR}/${f}.id.targets)
+
+       find ${OUTDIR} -name "${f}-${T}-split*.marc" \
+       -exec cat \{} \; \
+       >${OUTDIR}/${f}-${T}.marc
+}
+
+split_suffix() {
+       local input="$1" output_prefix="$2" output_suffix="$3" split_count="$4"
+       split -a 7 -d -C ${split_count} ${input} ${output_prefix}
+       find $(dirname ${output_prefix}) -name "$(basename ${output_prefix})*" -exec mv -f \{\} \{\}${output_suffix} \;
+}
+
+# media for BC ELN post-secondary libraries
+export SPLIT_COUNT=10000
+F=bc_eln_media SQL="$POSTSEC_QUERY"
+process_batch "$F" "${SQL}" $SPLIT_COUNT
+
+# serials for BC ELN post-secondary libraries
+export SPLIT_COUNT=10000
+F=bc_eln_serials SQL="$POSTSEC_QUERY"
+process_batch "$F" "${SQL}" $SPLIT_COUNT
+
+export SPLIT_COUNT=100000
+# full dump for outlook public libraries
+F=sitka_full_outlook SQL="$FULL_QUERY"
+process_batch "$F" "${SQL}" $SPLIT_COUNT
+