From a4784bbacc356d381ff833e1c181b75c30f776c6 Mon Sep 17 00:00:00 2001 From: Louie Shprung Date: Fri, 9 Jun 2023 11:26:54 -0700 Subject: Clean up, base more on dash-docset-template, dumb down entry types for name-index (consider all of these directives) --- src/index-pages.sh | 35 +++++++++++++++ src/index-terms.sh | 33 ++++++++++++++ src/index.sh | 122 --------------------------------------------------- src/lib/create_table | 7 +++ src/lib/insert | 8 ++++ 5 files changed, 83 insertions(+), 122 deletions(-) create mode 100755 src/index-pages.sh create mode 100755 src/index-terms.sh delete mode 100755 src/index.sh create mode 100644 src/lib/create_table create mode 100644 src/lib/insert (limited to 'src') diff --git a/src/index-pages.sh b/src/index-pages.sh new file mode 100755 index 0000000..99531b7 --- /dev/null +++ b/src/index-pages.sh @@ -0,0 +1,35 @@ +#!/usr/bin/env sh + +# shellcheck source=./lib/create_table +. "$(dirname "$0")"/lib/create_table +# shellcheck source=./lib/insert +. "$(dirname "$0")"/lib/insert + +DB_PATH="$1" +shift + +get_title() { + FILE="$1" + + pup -p -f "$FILE" 'title text{}' | \ + tr -d \\n | \ + sed 's/(GNU make)//g' | \ + sed 's/\"/\"\"/g' +} + +insert_pages() { + # Get title and insert into table for each html file + while [ -n "$1" ]; do + unset PAGE_NAME + unset PAGE_TYPE + PAGE_NAME="$(get_title "$1")" + PAGE_TYPE="Guide" + if [ -n "$PAGE_NAME" ]; then + insert "$DB_PATH" "$PAGE_NAME" "$PAGE_TYPE" "$(basename "$1")" + fi + shift + done +} + +create_table "$DB_PATH" +insert_pages "$@" diff --git a/src/index-terms.sh b/src/index-terms.sh new file mode 100755 index 0000000..f74af03 --- /dev/null +++ b/src/index-terms.sh @@ -0,0 +1,33 @@ +#!/usr/bin/env sh + +# shellcheck source=./lib/create_table +. "$(dirname "$0")"/lib/create_table +# shellcheck source=./lib/insert +. "$(dirname "$0")"/lib/insert + +TYPE="$1" +shift +DB_PATH="$1" +shift + +insert_index_terms() { + # Get each term from an index page and insert + while [ -n "$1" ]; do + grep -Eo ":" "$1" | while read -r line; do + insert_term "$line" + done + + shift + done +} + +insert_term() { + LINK="$1" + NAME="$(echo "$LINK" | pup -p 'a text{}' | sed 's/\"\"//g' | tr -d \\n)" + PAGE_PATH="$(echo "$LINK" | pup -p 'a attr{href}')" + + insert "$DB_PATH" "$NAME" "$TYPE" "$PAGE_PATH" +} + +create_table "$DB_PATH" +insert_index_terms "$@" diff --git a/src/index.sh b/src/index.sh deleted file mode 100755 index 3b4cacd..0000000 --- a/src/index.sh +++ /dev/null @@ -1,122 +0,0 @@ -#!/usr/bin/env sh - -create_table() { - sqlite3 "$DB_PATH" "CREATE TABLE IF NOT EXISTS searchIndex(id INTEGER PRIMARY KEY, name TEXT, type TEXT, path TEXT);" - sqlite3 "$DB_PATH" "CREATE UNIQUE INDEX IF NOT EXISTS anchor ON searchIndex (name, type, path);" -} - -get_title() { - FILE="$1" - - pup -p -f "$FILE" 'title text{}' | \ - tr -d \\n | \ - sed 's/(GNU make)//g' | \ - sed 's/\"/\"\"/g' -} - -get_type() { - LINK="$(echo "$1" | sed 's/#[^#]*$//')" - LINK_TITLE="$(get_title "$2/$LINK")" - - set -- $POSSIBLE_TYPES - - while [ -n "$1" ]; do - if echo "$LINK_TITLE" | grep -iq "$1"; then - echo "$1" - return - fi - shift - done - - echo "Entry" -} - -insert() { - NAME="$1" - TYPE="$2" - PAGE_PATH="$3" - - sqlite3 "$DB_PATH" "INSERT INTO searchIndex(name, type, path) VALUES (\"$NAME\",\"$TYPE\",\"$PAGE_PATH\");" -} - -insert_index_terms() { - # Get each term from an index page and insert - while [ -n "$1" ]; do - grep -Eo ":" "$1" | while read -r line; do - insert_term "$line" "$(dirname "$1")" - done - - shift - done -} - - -insert_pages() { - # Get title and insert into table for each html file - while [ -n "$1" ]; do - unset PAGE_NAME - unset PAGE_TYPE - PAGE_NAME="$(get_title "$1")" - if [ -n "$PAGE_NAME" ]; then - PAGE_TYPE="Guide" - insert "$PAGE_NAME" "$PAGE_TYPE" "$(basename "$1")" - fi - shift - done -} - -insert_term() { - LINK="$1" - PAGE_DIR="$2" - - NAME="$(echo "$LINK" | pup -p 'a text{}' | tr -d \\n | sed 's/"/\"\"/g')" - TYPE="$INDEX_TYPE" - PAGE_PATH="$(echo "$LINK" | pup -p 'a attr{href}')" - if [ -n "$POSSIBLE_TYPES" ]; then - TYPE="$(get_type "$PAGE_PATH" "$PAGE_DIR")" - elif [ -z "$TYPE" ]; then - TYPE="Entry" - fi - - insert "$NAME" "$TYPE" "$PAGE_PATH" -} - -TYPE="PAGES" - -# Check flags -while true; do - case "$1" in - -c|--check) - # List of space-separated possible index entry types (overwrites -t) - shift - POSSIBLE_TYPES="$1" - shift - ;; - -i|--index) - # Set the script to handle index pages - TYPE="INDEX" - shift - ;; - -t|--type) - # Set a type for index entries - shift - INDEX_TYPE="$1" - shift - ;; - *) - break - esac -done - -DB_PATH="$1" -shift - -create_table -case "$TYPE" in - PAGES) - insert_pages "$@" - ;; - INDEX) - insert_index_terms "$@" - ;; -esac diff --git a/src/lib/create_table b/src/lib/create_table new file mode 100644 index 0000000..a783c50 --- /dev/null +++ b/src/lib/create_table @@ -0,0 +1,7 @@ +create_table() { + DB_PATH="$1" + + sqlite3 "$DB_PATH" "CREATE TABLE IF NOT EXISTS searchIndex(id INTEGER PRIMARY KEY, name TEXT, type TEXT, path TEXT);" + sqlite3 "$DB_PATH" "CREATE UNIQUE INDEX IF NOT EXISTS anchor ON searchIndex (name, type, path);" +} + diff --git a/src/lib/insert b/src/lib/insert new file mode 100644 index 0000000..31c1b4c --- /dev/null +++ b/src/lib/insert @@ -0,0 +1,8 @@ +insert() { + DB_PATH="$1" + NAME="$2" + TYPE="$3" + PAGE_PATH="$4" + + sqlite3 "$DB_PATH" "INSERT INTO searchIndex(name, type, path) VALUES (\"$NAME\",\"$TYPE\",\"$PAGE_PATH\");" +} -- cgit