diff options
| author | sigoden <sigoden@gmail.com> | 2024-07-06 02:45:39 +0000 |
|---|---|---|
| committer | sigoden <sigoden@gmail.com> | 2024-07-06 02:45:39 +0000 |
| commit | 9e23134b7cc583538f2b9667ffa251b575985bc6 (patch) | |
| tree | 674e1cb34d2dc5fc4a264fea9f79fa194f5bde2e /tools/get_web_page.sh | |
| parent | 02067cd913b53fd04b9d7287a396825c65092ae8 (diff) | |
| download | llm-functions-docker-9e23134b7cc583538f2b9667ffa251b575985bc6.tar.gz | |
refactor: improve code quality
Diffstat (limited to 'tools/get_web_page.sh')
| -rwxr-xr-x | tools/get_web_page.sh | 9 |
1 files changed, 5 insertions, 4 deletions
diff --git a/tools/get_web_page.sh b/tools/get_web_page.sh index 65e6d2a..4e1e16a 100755 --- a/tools/get_web_page.sh +++ b/tools/get_web_page.sh @@ -1,14 +1,15 @@ #!/usr/bin/env bash set -e -# @describe Fetches the HTML content from a specified webpage URL and converts it to Markdown format. -# Use it to answer user questions that require up-to-date content from web pages. -# @meta require-tools curl,pandoc,sed +# @describe Get the contents from a URL. +# @meta require-tools pandoc # @option --url! The URL to scrape. main() { # span and div tags are dropped from the HTML https://pandoc.org/MANUAL.html#raw-htmltex and sed removes any inline SVG images in image tags from the Markdown content. - curl -fsSL "$argc_url" | pandoc -f html-native_divs-native_spans -t gfm-raw_html | sed -E 's/!\[.*?\]\((data:image\/svg\+xml[^)]+)\)//g' + curl -fsSL "$argc_url" | \ + pandoc -f html-native_divs-native_spans -t gfm-raw_html | \ + sed -E 's/!\[.*?\]\((data:image\/svg\+xml[^)]+)\)//g' } eval "$(argc --argc-eval "$0" "$@")" |
