thatonelutenist/website
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Description generation

Browse source
This commit is contained in:
Nathan McCarty 2025-02-04 16:51:37 -05:00
parent 804b51aaa6
commit ce364c01d8
6 changed files with 102 additions and 33 deletions
Download patch file Download diff file Expand all files Collapse all files

69
lib/Pandoc.rakumod
View file

@ -3,13 +3,10 @@ unit module Pandoc;
use JSON::Fast;
#| Extract the title from a markdown document
#|
#| The title is the only top level header, will throw an error if there are
#| multiple top level headers or are none
sub markdown-title(IO::Path:D $file --> Str:D) is export {
#| Run pandoc with the given arguments, dieing on failure
sub pandoc(*@args --> Str:D) {
# Call into pandoc
my $pandoc = run <pandoc -f gfm -t JSON>, $file, :out, :err;
my $pandoc = run 'pandoc', @args, :out, :err;
# Collect the output
my $output = $pandoc.out.slurp: :close;
@ -17,6 +14,17 @@ sub markdown-title(IO::Path:D $file --> Str:D) is export {
die "Pandoc exited with {$pandoc.exitcode}\nout: $output\nerr: $stderr"
unless $pandoc;
$output
}
#| Extract the title from a markdown document
#|
#| The title is the only top level header, will throw an error if there are
#| multiple top level headers or are none
sub markdown-title(IO::Path:D $file --> Str:D) is export {
# Collect the output
my $output = pandoc <-f gfm -t JSON>, $file;
# Parse out output from pandoc, we are making an executive decision to trust
# pandoc here, so we won't do any error handling for pandoc's output
my %parsed = from-json $output;
@ -27,12 +35,8 @@ sub markdown-title(IO::Path:D $file --> Str:D) is export {
$v ~~ Associative && $v<t> ~~ "Header"
}
my @headers = %parsed<blocks>.grep(&is-header).grep(*<c>[0] == 1);
if @headers.elems > 1 {
die "More than one top level header in $file";
};
if @headers.elems == 0 {
die "No top level headers in $file";
};
die "More than one top level header in $file" if @headers.elems > 1;
die "No top level headers in $file" if @headers.elems == 0;
# Extract the header and process it into a string
my @header = @headers[0]<c>[2].flat;
@ -55,15 +59,38 @@ sub markdown-title(IO::Path:D $file --> Str:D) is export {
return $title;
}
#| Use pandoc to extract the first paragraph of a markdown document
sub markdown-first-paragraph(IO::Path:D $file --> Str:D) is export {
my $output = pandoc <-f gfm -t JSON>, $file;
my %parsed = from-json $output;
# Extract a list of paragraphs from the pandoc output
my sub is-para($v) {
$v ~~ Associative && $v<t> ~~ 'Para'
}
my @paras = %parsed<blocks>.grep(&is-para);
die "No paragraphs in markdown" if @paras.elems == 0;
my @para = @paras[0][0]<c>.flat;
# Proces it into a string
my $para = "";
for @para -> $component {
next unless $component ~~ Associative;
given $component<t> {
when "Str" {
$para ~= $component<c>;
}
when "Space" {
$para ~= " ";
}
default {
die "Invalid component type: $_";
}
}
}
$para
}
#| Use pandoc to render a markdown document to html
sub markdown-to-html(IO::Path:D $file --> Str:D) is export {
# Call into pandoc
my $pandoc = run <pandoc -f gfm>, $file, :out, :err;
# Collect the output
my $output = $pandoc.out.slurp: :close;
my $stderr = $pandoc.err.slurp: :close;
die "Pandoc exited with {$pandoc.exitcode}\nout: $output\nerr: $stderr"
unless $pandoc;
$output
pandoc <-f gfm>, $file
}