commit c57133401dc4cfd4b5eeb5bfedcd0dec76497c4e
parent 822d58f36aef387dfed8975226071c2fcf6c86d1
Author: JayVii <jayvii[AT]posteo[DOT]de>
Date: Fri, 11 Apr 2025 22:54:26 +0200
feat: uniformly using English; introducing optarg
Diffstat:
| M | Makefile | | | 5 | ++++- |
| M | gen_opml.sh | | | 88 | ++++++++++++++++++++++++++++++++++++++++++++++++++++++------------------------- |
2 files changed, 64 insertions(+), 29 deletions(-)
diff --git a/Makefile b/Makefile
@@ -1,5 +1,8 @@
+# SPDX-License-Identifier: AGPL-3.0-or-later
+# SPDX-FileCopyrightText: 2025 JayVii <jayvii[AT]posteo[DOT]de>
+
gen_opml:
mkdir -p ./opml # ensure directory exists
rm -r ./opml
mkdir -p ./opml
- ./gen_opml.sh
+ ./gen_opml.sh -s quellen.tsv
diff --git a/gen_opml.sh b/gen_opml.sh
@@ -1,9 +1,33 @@
#!/usr/bin/env bash
+# SPDX-License-Identifier: AGPL-3.0-or-later
+# SPDX-FileCopyrightText: 2025 JayVii <jayvii[AT]posteo[DOT]de>
+
+# Gather input -----------------------------------------------------------------
+
+# fetch source file from the "-s " flag
+while getopts "s:" opt; do
+ case $opt in
+ s)
+ export sources_file="$OPTARG"
+ ;;
+ esac
+done
+
+# exit if sourcefile is not provided
+if [ -z $sources_file ]; then
+ echo "Usage: $0 -s /path/to/sources.tsv" > /dev/stderr
+ exit 1
+fi
+
# Functions --------------------------------------------------------------------
function parse_tsv_line {
+
+ # escape input as replacement string for regex
local repl="\\${2}"
+
+ # split up single-line string into its four parts and return according part
echo "$1" | \
tail -n 1 | \
sed -E -e "s/^([^\t]+)\t+([^\t]+)\t+([^\t]+)\t+([^\t]+)$/${repl}/"
@@ -11,25 +35,30 @@ function parse_tsv_line {
# Script -----------------------------------------------------------------------
+# User output
+echo "Gather information from ${sources_file} ..."
+
# parse TSV file and assign to arrays
while read -r line; do
name_tmp=$(parse_tsv_line "$line" 1)
- katg_tmp=$(parse_tsv_line "$line" 2)
+ catg_tmp=$(parse_tsv_line "$line" 2)
addr_tmp=$(parse_tsv_line "$line" 3)
rssa_tmp=$(parse_tsv_line "$line" 4)
name+=("$name_tmp")
- katg+=("$katg_tmp")
+ catg+=("$catg_tmp")
addr+=("$addr_tmp")
rssa+=("$rssa_tmp")
-done < quellen.tsv
+done < ${sources_file}
# create an array of unique categories
-for i in $(seq 1 1 ${#katg[@]}); do
- katg_liste+="$(echo ${katg[$i]} | sed -e 's/\s/_/g' -e 's/,/\ /g') "
+## create space separated single string of categories
+for i in $(seq 1 1 ${#catg[@]}); do
+ catg_list+="$(echo ${catg[$i]} | sed -e 's/\s/_/g' -e 's/,/\ /g') "
done
-katg_liste=($(echo "$katg_liste" | tr ' ' '\n' | sort -u | tr '\n' ' '))
+## sort and return only unique categories into array
+catg_list=($(echo "$catg_list" | tr ' ' '\n' | sort -u | tr '\n' ' '))
# define OPML header
opml_header="<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n"
@@ -39,25 +68,28 @@ opml_header+=" <title>LinkAgg Newsaggregator</title>\n"
opml_header+=" <dateCreated>$(LC_ALL=en date)</dateCreated>\n"
opml_header+=" </head>\n"
-# build OPML file kategory by kategory
-for k in ${katg_liste[@]}; do
+# build OPML file category-by-category
+for cat in ${catg_list[@]}; do
- k=$(echo "$k" | tr '_' ' ')
- echo "Bearbeite: ${k}..."
+ # re-introduce spaces into categories again
+ cat=$(echo "$cat" | tr '_' ' ')
- # Start der Kategorien "outline"
- opml_body=" <outline title=\"$k\" text=\"$k\">\n"
+ # output for user
+ echo "Generating ./opml/${cat}.opml"
- # Schleife durch alle Quellen
+ # First part of opml body definition (category outline)
+ opml_body=" <outline title=\"$cat\" text=\"$cat\">\n"
+
+ # Loop through each source individually
for i in $(seq 1 1 ${#name[@]}); do
- # Test ob Kategorie in Quelle enthalten ist
- if [ $(echo ${katg[$i]} | grep -c -E "(^|,)${k}($|,)") -gt 0 ]; then
+ # If source contains the category, add it to the OPML body
+ if [ $(echo ${catg[$i]} | grep -c -E "(^|,)${cat}($|,)") -gt 0 ]; then
- # RSS-Eintrag für zutreffende Quelle
+ # generate OPML line for the current source
a="title=\"${name[$i]}\""
b="text=\"${name[$i]}\""
- c="category=\"$(echo ${katg[$i]} | sed -E -e 's/(^|,)/\1\//g')\""
+ c="category=\"$(echo ${catg[$i]} | sed -E -e 's/(^|,)/\1\//g')\""
d="htmlUrl=\"${addr[$i]}\""
e="xmlUrl=\"${rssa[$i]}\""
opml_body+=" <outline $a $b $c $d $e type=\"rss\" />\n"
@@ -66,24 +98,24 @@ for k in ${katg_liste[@]}; do
done
- # Ende der Kategorie "outline"
+ # Last part of opml body definition
opml_body+=" </outline>\n"
- # Übernahme in globalen OPML-Körper
+ # Copy current category opml body to the global string variable
opml_body_global+="$opml_body"
- # Export der kategorien OPML-Datei
- kat_header=$(
+ # export current category into its own category file
+ cat_header=$(
echo "$opml_header" | sed -E -e "s/(<\/title>)/Kategorie: ${k}\1/"
)
- kat_opml="${kat_header} <body>\n${opml_body} </body>\n</opml>"
- printf "${kat_opml}\n" | tee "./opml/${k}.opml" > /dev/null
+ cat_opml="${cat_header} <body>\n${opml_body} </body>\n</opml>"
+ printf "${cat_opml}\n" | tee "./opml/${cat}.opml" > /dev/null
done
-# Export der globalen OPML-Datei
-opml_head=" <head>\n"
-opml_head+=" <title>LinkAgg Newsaggregator</title>\n"
-opml_head+=" </head>\n"
+# Export of global OPML file
+echo "Generating ./opml/linkagg.opml"
opml="${opml_header}${opml_head} <body>\n${opml_body_global} </body>\n</opml>"
-printf "${opml}\n" | tee "./opml/linkagg.opml.xml" > /dev/null
+printf "${opml}\n" | tee "./opml/linkagg.opml" > /dev/null
+
+echo "Done!"
