Try to parallelize RSS building better

This commit is contained in:
Yann Esposito (Yogsototh) 2021-05-07 18:02:58 +02:00
parent 1e727b5d2b
commit 2252391159
Signed by untrusted user who does not match committer: yogsototh
GPG Key ID: 7B19A4C650D59646
3 changed files with 104 additions and 43 deletions

View File

@ -7,6 +7,7 @@
all: fast
SRC_DIR ?= src
DST_DIR ?= _site
CACHE_DIR ?= .cache
# we don't want to publish files in drafts
NO_DRAFT := -not -path '$(SRC_DIR)/drafts/*'
@ -44,6 +45,8 @@ $(DST_DIR)/%.html: $(SRC_DIR)/%.org $(TEMPLATE)
@rm $@.tmp
ALL += $(DST_PANDOC_FILES)
# HTML INDEX
HTML_INDEX := $(DST_DIR)/index.html
MKINDEX := engine/mk-index.sh
@ -53,9 +56,21 @@ $(HTML_INDEX): $(DST_PANDOC_FILES) $(MKINDEX)
ALL += $(HTML_INDEX)
# RSS
SRC_POSTS_DIR ?= $(SRC_DIR)/posts
SRC_POST_FILES ?= $(shell find $(SRC_POSTS_DIR) -type f -name "*$(EXT)")
RSS_CACHE_DIR ?= $(CACHE_DIR)/rss
DST_RSS_FILES ?= $(subst .$(EXT),.rss, \
$(patsubst $(SRC_POSTS_DIR)/%,$(RSS_CACHE_DIR)/%, \
$(SRC_POSTS_FILES)))
MK_RSS_ENTRY := ./engine/mk-rss-entry.sh
$(RSS_CACHE_DIR)/%.rss: $(DST_DIR)/posts/%.html $(MK_RSS_ENTRY)
@mkdir -p $(RSS_CACHE_DIR)
$(MK_RSS_ENTRY) $@
RSS := $(DST_DIR)/rss.xml
MKRSS := engine/mkrss.sh
$(RSS): $(DST_PANDOC_FILES) $(MKRSS)
$(RSS): $(DST_RSS_FILES) $(MKRSS)
$(MKRSS)
ALL += $(RSS)
@ -119,3 +134,4 @@ fast: $(ALL)
clean:
-[ ! -z "$(DST_DIR)" ] && rm -rf $(DST_DIR)/*
-[ ! -z "$(CACHE_DIR)" ] && rm -rf $(CACHE_DIR)/*

72
engine/mk-rss-entry.sh Executable file
View File

@ -0,0 +1,72 @@
#!/usr/bin/env zsh
cd "$(git rev-parse --show-toplevel)" || exit 1
# Directory
webdir="_site"
postsdir="$webdir/posts"
indexdir=".cache/rss"
# file to handle
fic="$1"
# RSS Metas
websiteurl="https://her.esy.fun"
# HTML Accessors (similar to CSS accessors)
dateaccessor='.yyydate'
contentaccessor='#content'
# title and keyword shouldn't be changed
titleaccessor='title'
keywordsaccessor='meta[name=keywords]::attr(content)'
formatdate() {
# format the date for RSS
local d="$1"
# echo "DEBUG DATE: $d" >&2
LC_TIME=en_US date --date $d +'%a, %d %b %Y %H:%M:%S %z'
}
finddate(){ < $1 hxselect -c $dateaccessor | sed 's/\[//g;s/\]//g;s/ .*$//' }
findtitle(){ < $1 hxselect -c $titleaccessor }
getcontent(){
< $1 hxselect $contentaccessor | \
perl -pe 'use URI; $base="'$2'"; s# (href|src)="((?!https?://)[^"]*)"#" ".$1."=\"".URI->new_abs($2,$base)->as_string."\""#eig' }
findkeywords(){ < $1 hxselect -c $keywordsaccessor | sed 's/,/ /g' }
mkcategories(){
for keyword in $*; do
printf "\\n<category>%s</category>" $keyword
done
}
autoload -U colors && colors
if echo $fic|egrep -- '-(mk|min|sci|modern).html$'>/dev/null; then
continue
fi
postfile="$(echo "$fic"|sed 's#^'$postsdir'/##')"
blogfile="$(echo "$fic"|sed 's#^'$webdir'/##')"
printf "%-30s" $postfile
xfic="$indexdir/$fic.xml"
mkdir -p $(dirname $xfic)
hxclean $fic > $xfic
d=$(finddate $xfic)
echo -n " [$d]"
rssdate=$(formatdate $d)
title=$(findtitle $xfic)
keywords=( $(findkeywords $xfic) )
printf ": %-55s" "$title ($keywords)"
categories=$(mkcategories $keywords)
absoluteurl="${websiteurl}/${blogfile}"
dst="$indexdir/$fic.rss"
mkdir -p $(dirname $dst)
{ printf "\\n<item>"
printf "\\n<title>%s</title>" "$title"
printf "\\n<guid>%s</guid>" "$absoluteurl"
printf "\\n<pubDate>%s</pubDate>%s" "$rssdate"
printf "%s" "$categories"
printf "\\n<description><![CDATA[\\n%s\\n]]></description>" "$(getcontent "$xfic" "$absoluteurl")"
printf "\\n</item>\\n\\n"
} >> "$dst"
echo " [${fg[green]}OK${reset_color}]"

View File

@ -5,6 +5,7 @@ cd "$(git rev-parse --show-toplevel)" || exit 1
webdir="_site"
postsdir="$webdir/posts"
rssfile="$webdir/rss.xml"
indexdir=".cache/rss"
# maximal number of articles to put in the RSS file
maxarticles=10
@ -19,11 +20,7 @@ rssauthor="yann@esposito.host (Yann Esposito)"
rssimgurl="https://her.esy.fun/img/FlatAvatar.png"
# HTML Accessors (similar to CSS accessors)
dateaccessor='.yyydate'
contentaccessor='#content'
# title and keyword shouldn't be changed
titleaccessor='title'
keywordsaccessor='meta[name=keywords]::attr(content)'
dateaccessor='pubDate'
formatdate() {
# format the date for RSS
@ -32,51 +29,27 @@ formatdate() {
LC_TIME=en_US date --date $d +'%a, %d %b %Y %H:%M:%S %z'
}
finddate(){ < $1 hxselect -c $dateaccessor | sed 's/\[//g;s/\]//g;s/ .*$//' }
findtitle(){ < $1 hxselect -c $titleaccessor }
getcontent(){
< $1 hxselect $contentaccessor | \
perl -pe 'use URI; $base="'$2'"; s# (href|src)="((?!https?://)[^"]*)"#" ".$1."=\"".URI->new_abs($2,$base)->as_string."\""#eig' }
findkeywords(){ < $1 hxselect -c $keywordsaccessor | sed 's/,/ /g' }
mkcategories(){
for keyword in $*; do
printf "\\n<category>%s</category>" $keyword
done
isodate() {
# format the date for sorting
local d="$1"
# echo "DEBUG DATE: $d" >&2
LC_TIME=en_US date --date $d +'%Y-%m-%dT%H:%M:%S'
}
finddate(){ < $1 hxselect -c $dateaccessor | sed 's/\[//g;s/\]//g;s/ .*$//' }
autoload -U colors && colors
tmpdir=$(mktemp -d)
typeset -a dates
dates=( )
for fic in $postsdir/**/*.html; do
if echo $fic|egrep -- '-(mk|min|sci|modern).html$'>/dev/null; then
continue
fi
postfile="$(echo "$fic"|sed 's#^'$postsdir'/##')"
blogfile="$(echo "$fic"|sed 's#^'$webdir'/##')"
printf "%-30s" $postfile
xfic="$tmpdir/$fic.xml"
mkdir -p $(dirname $xfic)
hxclean $fic > $xfic
d=$(finddate $xfic)
echo -n " [$d]"
rssdate=$(formatdate $d)
title=$(findtitle $xfic)
keywords=( $(findkeywords $xfic) )
printf ": %-55s" "$title ($keywords)"
categories=$(mkcategories $keywords)
absoluteurl="${websiteurl}/${blogfile}"
{ printf "\\n<item>"
printf "\\n<title>%s</title>" "$title"
printf "\\n<guid>%s</guid>" "$absoluteurl"
printf "\\n<pubDate>%s</pubDate>%s" "$rssdate"
printf "%s" "$categories"
printf "\\n<description><![CDATA[\\n%s\\n]]></description>" "$(getcontent "$xfic" "$absoluteurl")"
printf "\\n</item>\\n\\n"
} >> "$tmpdir/${d}-$(basename $fic).rss"
tmpdir=$(mktemp -d)
for fic in $indexdir/*.rss; do
rssdate=$(finddate $xfic)
echo -n "${fic:t} [$d]"
d=$(isodate $rssdate)
dates=( $d $dates )
echo " [${fg[green]}OK${reset_color}]"
cp $fic $tmpdir/$d-${fic:t}.rss
done
echo "Publishing"
for fic in $(ls $tmpdir/*.rss | sort -r | head -n $maxarticles ); do