1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178

//! Extract author information for book clusters.
use std::path::PathBuf;

use parse_display::{Display, FromStr};

use crate::arrow::dfext::*;
use crate::arrow::writer::open_parquet_writer;
use crate::io::object::ThreadObjectWriter;
use crate::prelude::*;
use crate::util::logging::data_progress;
use anyhow::Result;
use polars::prelude::*;

#[derive(Display, FromStr, Debug, Clone)]
#[display(style = "lowercase")]
enum Source {
    OpenLib,
    LOC,
}

#[derive(Args, Debug)]
#[command(name = "extract-authors")]
/// Extract cluster author data from extracted book data.
pub struct ClusterAuthors {
    /// Only extract first authors
    #[arg(long = "first-author")]
    first_author: bool,

    /// Specify output file
    #[arg(short = 'o', long = "output")]
    output: PathBuf,

    /// Specify the source
    #[arg(short = 's', long = "source")]
    sources: Vec<Source>,
}

/// Scan the OpenLibrary data for authors.
fn scan_openlib(first_only: bool) -> Result<LazyFrame> {
    info!("scanning OpenLibrary author data");
    info!("reading ISBN clusters");
    let icl = scan_df_parquet("book-links/isbn-clusters.parquet")?;
    let icl = icl.select(&[col("isbn_id"), col("cluster")]);
    info!("reading OL edition IDs");
    let edl = scan_df_parquet("openlibrary/edition-isbn-ids.parquet")?;
    let edl = edl.filter(col("isbn_id").is_not_null());
    info!("reading OL edition authors");
    let mut eau = scan_df_parquet("openlibrary/edition-authors.parquet")?;
    if first_only {
        eau = eau.filter(col("pos").eq(0i16));
    }

    info!("reading OL author names");
    let auth = scan_df_parquet("openlibrary/author-names.parquet")?;
    let linked = icl.join(
        edl,
        [col("isbn_id")],
        [col("isbn_id")],
        JoinType::Inner.into(),
    );
    let linked = linked.join(
        eau,
        [col("edition")],
        [col("edition")],
        JoinType::Inner.into(),
    );
    let linked = linked.join(auth, [col("author")], [col("id")], JoinType::Inner.into());
    let authors = linked.select(vec![
        col("cluster"),
        col("name")
            .alias("author_name")
            .map(udf_clean_name, GetOutput::from_type(DataType::String)),
    ]);

    Ok(authors)
}

/// Scan the Library of Congress data for first authors.
fn scan_loc(first_only: bool) -> Result<LazyFrame> {
    if !first_only {
        error!("only first-author extraction is currently supported");
        return Err(anyhow!("cannot extract multiple authors"));
    }

    info!("reading ISBN clusters");
    let icl = scan_df_parquet("book-links/isbn-clusters.parquet")?;
    let icl = icl.select([col("isbn_id"), col("cluster")]);

    info!("reading LOC book records");
    let books = scan_df_parquet("loc-mds/book-isbn-ids.parquet")?;

    info!("reading LOC book authors");
    let authors = scan_df_parquet("loc-mds/book-authors.parquet")?;
    let authors = authors.filter(col("author_name").is_not_null());

    let linked = icl.join(
        books,
        [col("isbn_id")],
        [col("isbn_id")],
        JoinType::Inner.into(),
    );
    let linked = linked.join(
        authors,
        [col("rec_id")],
        [col("rec_id")],
        JoinType::Inner.into(),
    );
    let authors = linked.select(vec![
        col("cluster"),
        col("author_name").map(udf_clean_name, GetOutput::from_type(DataType::String)),
    ]);

    Ok(authors)
}

impl Command for ClusterAuthors {
    fn exec(&self) -> Result<()> {
        let mut authors: Option<LazyFrame> = None;
        for source in &self.sources {
            let astr = match source {
                Source::OpenLib => scan_openlib(self.first_author)?,
                Source::LOC => scan_loc(self.first_author)?,
            };
            debug!("author source {} has schema {:?}", source, astr.schema());
            if let Some(adf) = authors {
                authors = Some(concat(
                    [adf, astr],
                    UnionArgs {
                        parallel: true,
                        rechunk: false,
                        to_supertypes: false,
                    },
                )?);
            } else {
                authors = Some(astr);
            }
        }
        let authors = authors.ok_or(anyhow!("no sources specified"))?;
        let authors = authors.filter(
            col("author_name")
                .is_not_null()
                .and(col("author_name").neq("".lit())),
        );

        let authors = authors.unique(None, UniqueKeepStrategy::First);

        debug!("plan: {}", authors.describe_plan());

        info!("collecting results");
        let authors = authors.collect()?;
        info!("found {} cluster-author links", authors.height());

        info!("saving to {:?}", &self.output);
        // clean up nullability
        // we do the writing ourself because we have no nulls, but polars doesn't deal with that
        let schema = nonnull_schema(&authors);
        debug!("schema: {:?}", schema);

        let writer = open_parquet_writer(&self.output, schema)?;
        let mut writer = ThreadObjectWriter::wrap(writer)
            .with_name("author parquet")
            .spawn();
        let pb = data_progress(authors.n_chunks());
        for chunk in authors.iter_chunks(false) {
            writer.write_object(chunk)?;
            pb.tick();
        }
        writer.finish()?;
        std::mem::drop(pb);

        info!(
            "output file is {}",
            friendly::bytes(file_size(&self.output)?)
        );

        Ok(())
    }
}