Merge pull request #109 from Kerollmops/implement-text-cropping

Introduce text cropping that shows the first matches
This commit is contained in:
Clément Renault 2019-02-18 19:40:30 +01:00 committed by GitHub
commit f5992ce822
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

View File

@ -28,6 +28,10 @@ pub struct Opt {
/// The number of returned results /// The number of returned results
#[structopt(short = "n", long = "number-results", default_value = "10")] #[structopt(short = "n", long = "number-results", default_value = "10")]
pub number_results: usize, pub number_results: usize,
/// The number of characters before and after the first match
#[structopt(short = "C", long = "context", default_value = "35")]
pub char_context: usize,
} }
type Document = HashMap<String, String>; type Document = HashMap<String, String>;
@ -67,12 +71,10 @@ fn char_to_byte_range(index: usize, length: usize, text: &str) -> (usize, usize)
(byte_index, byte_length) (byte_index, byte_length)
} }
fn create_highlight_areas(text: &str, matches: &[Match], attribute: SchemaAttr) -> Vec<usize> { fn create_highlight_areas(text: &str, matches: &[Match]) -> Vec<usize> {
let mut byte_indexes = BTreeMap::new(); let mut byte_indexes = BTreeMap::new();
for match_ in matches { for match_ in matches {
let match_attribute = match_.attribute;
if SchemaAttr::new(match_attribute) == attribute {
let char_index = match_.char_index as usize; let char_index = match_.char_index as usize;
let char_length = match_.char_length as usize; let char_length = match_.char_length as usize;
let (byte_index, byte_length) = char_to_byte_range(char_index, char_length, text); let (byte_index, byte_length) = char_to_byte_range(char_index, char_length, text);
@ -86,7 +88,6 @@ fn create_highlight_areas(text: &str, matches: &[Match], attribute: SchemaAttr)
}, },
} }
} }
}
let mut title_areas = Vec::new(); let mut title_areas = Vec::new();
title_areas.push(0); title_areas.push(0);
@ -99,6 +100,39 @@ fn create_highlight_areas(text: &str, matches: &[Match], attribute: SchemaAttr)
title_areas title_areas
} }
/// note: matches must have been sorted by `char_index` and `char_length` before being passed.
///
/// ```no_run
/// matches.sort_unstable_by_key(|m| (m.char_index, m.char_length));
///
/// let matches = matches.matches.iter().filter(|m| SchemaAttr::new(m.attribute) == attr).cloned();
///
/// let (text, matches) = crop_text(&text, matches, 35);
/// ```
fn crop_text(
text: &str,
matches: impl IntoIterator<Item=Match>,
context: usize,
) -> (String, Vec<Match>)
{
let mut matches = matches.into_iter().peekable();
let char_index = matches.peek().map(|m| m.char_index as usize).unwrap_or(0);
let start = char_index.saturating_sub(context);
let text = text.chars().skip(start).take(context * 2).collect();
let matches = matches
.take_while(|m| {
(m.char_index as usize) + (m.char_length as usize) <= start + (context * 2)
})
.map(|match_| {
Match { char_index: match_.char_index - start as u32, ..match_ }
})
.collect();
(text, matches)
}
fn main() -> Result<(), Box<Error>> { fn main() -> Result<(), Box<Error>> {
let _ = env_logger::init(); let _ = env_logger::init();
let opt = Opt::from_args(); let opt = Opt::from_args();
@ -126,7 +160,10 @@ fn main() -> Result<(), Box<Error>> {
let documents = builder.query(query, 0..opt.number_results); let documents = builder.query(query, 0..opt.number_results);
let number_of_documents = documents.len(); let number_of_documents = documents.len();
for doc in documents { for mut doc in documents {
doc.matches.sort_unstable_by_key(|m| (m.char_index, m.char_index));
match view.document_by_id::<Document>(doc.id) { match view.document_by_id::<Document>(doc.id) {
Ok(document) => { Ok(document) => {
for name in &opt.displayed_fields { for name in &opt.displayed_fields {
@ -140,7 +177,11 @@ fn main() -> Result<(), Box<Error>> {
}; };
print!("{}: ", name); print!("{}: ", name);
let areas = create_highlight_areas(&text, &doc.matches, attr); let matches = doc.matches.iter()
.filter(|m| SchemaAttr::new(m.attribute) == attr)
.cloned();
let (text, matches) = crop_text(&text, matches, opt.char_context);
let areas = create_highlight_areas(&text, &matches);
display_highlights(&text, &areas)?; display_highlights(&text, &areas)?;
println!(); println!();
} }