aboutsummaryrefslogtreecommitdiff
path: root/src/vector.rs
blob: c85c50a7e490027fdee7fc9611d4b62916c36b98 (plain)
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
use std::collections::HashMap;
use std::hash::Hash;
use std::ops::{Deref, DerefMut};
use serde::{Deserialize, Serialize};
use hash32::Hasher;

pub type Count = u8;
pub type Indexer = u32;

/// Represents the content of a cached file.
/// It is stored as a HashMap, because we do not
/// have to store the zeros. With that we save a lot
/// of storage.
#[derive(Default, Clone, Debug, Deserialize, Serialize)]
pub struct FileVector {
    data : Vec<FileVectorEntry>
}

#[repr(packed)]
#[derive(Default, Clone, Debug, Deserialize, Serialize)]
pub struct FileVectorEntry {
    index: u32,
    count: Count,
}

impl Deref for FileVector {
    type Target = Vec<FileVectorEntry>;
    fn deref(&self) -> &Self::Target {
        &self.data
    }
}

impl DerefMut for FileVector {
    fn deref_mut(&mut self) -> &mut Self::Target {
        &mut self.data
    }
}

impl FileVector {
    pub fn new() -> Self {
        Self { data : Vec::new() }
    }

    pub fn to_hashmap(&self) -> HashMap<Indexer, Count> {
        let mut map = HashMap::new();

        for e in self.data.iter() {
            map.insert(e.index, e.count);
        }

        map
    }

    pub fn from_words(words: Vec<String>) -> Self {
        let mut data : HashMap<Indexer, Count> = HashMap::new();

        for word in words {
            let mut hasher = hash32::FnvHasher::default();
            word.hash(&mut hasher);
            let k = hasher.finish32();
            match data.entry(k as Indexer) {
                std::collections::hash_map::Entry::Occupied(mut e) => {
                    let i = *e.get();
                    if  i == Count::MAX {
                        e.insert((i + 1) as Count);
                    }
                }
                std::collections::hash_map::Entry::Vacant(e) => {
                    e.insert(1);
                }
            }
        }

        FileVector::from_hashmap(data)
    }

    pub fn from_hashmap(map : HashMap<Indexer, Count>) -> Self {
        Self { data : Vec::from_iter(map.iter().map(|e| {
            FileVectorEntry { index: *e.0, count: *e.1 }
        }))}
    }
}

pub fn scalar_product(a : &FileVector, b : &FileVector) -> u32 {
    let a = a.to_hashmap();
    let b = b.to_hashmap();
    let mut c : u32 = 0;
    for (i, x) in a.iter() {
        c += (x * (b.get(i).unwrap_or(&0))) as u32;
    }
    c
}

pub fn match_vector(query : &FileVector, v : &FileVector) -> u32 {
    let query = query.to_hashmap();
    let v = v.to_hashmap();
    let mut c : u32 = 0;
    for (i, x) in query.iter() {
        let s = x * (v.get(i).unwrap_or(&0));
        if s == 0 {
            return 0
        } else {
            c += s as u32;
        }
    }
    c
}