pubmed-download-HOWTO

mkdir pubmed-scrape
cd pubmed-scrape
wget http://pmc.jensenlab.org/pmcid

grep nxml pmcid |sed 's/.*href="//;s/".*//;s/^/http:\/\/pmc.jensenlab.org\/pmcid\//' > xml-urls
for i in `seq -w 0 99`; do grep "$i\\.nxml" xml-urls > xml-urls-$i; done
mkdir xml
pushd xml
killall -9 xargs; killall wget
for i in ../xml-urls-*; do (num=${i#../xml-urls-}; mkdir -p $num && cd $num && xargs wget -qc < <((cat ../$i ; ls |sed 's,^,http://pmc.jensenlab.org/pmcid/,')|sort|uniq -u) )& done
popd

grep tar.gz pmcid |sed 's/.*href="//;s/".*//;s/^/http:\/\/pmc.jensenlab.org\/pmcid\//' > targz-urls
for i in `seq -w 0 99`; do grep "$i\\.tar\\.gz" targz-urls > targz-urls-$i; done
mkdir targz
pushd targz
killall -9 xargs; killall wget
for i in ../targz-urls-*; do (num=${i#../targz-urls-}; mkdir -p $num && cd $num && xargs wget -qc < <((cat ../$i ; ls |sed 's,^,http://pmc.jensenlab.org/pmcid/,')|sort|uniq -u) )& done
popd
popd