feat(schedule)!: move schedule parser, downloader, and updater to external library

This can be used to support more schedule formats in the future.

providers/provider-engels-polytechnic/src/lib.rs

@@ -0,0 +1,85 @@
+use crate::updater::Updater;
+use async_trait::async_trait;
+use base::{ScheduleProvider, ScheduleSnapshot};
+use std::ops::DerefMut;
+use std::sync::Arc;
+use std::time::Duration;
+use tokio::sync::RwLock;
+use tokio::time::interval;
+use tokio_util::sync::CancellationToken;
+
+pub use crate::updater::UpdateSource;
+
+mod parser;
+mod updater;
+mod xls_downloader;
+
+#[cfg(feature = "test")]
+pub mod test_utils {
+    pub use crate::parser::test_utils::test_result;
+}
+
+pub struct EngelsPolytechnicProvider {
+    updater: Updater,
+    snapshot: Arc<ScheduleSnapshot>,
+}
+
+impl EngelsPolytechnicProvider {
+    pub async fn new(
+        update_source: UpdateSource,
+    ) -> Result<Arc<dyn ScheduleProvider>, crate::updater::error::Error> {
+        let (updater, snapshot) = Updater::new(update_source).await?;
+
+        Ok(Arc::new(Wrapper {
+            inner: RwLock::new(Self {
+                updater,
+                snapshot: Arc::new(snapshot),
+            }),
+        }))
+    }
+}
+
+pub struct Wrapper {
+    inner: RwLock<EngelsPolytechnicProvider>,
+}
+
+#[async_trait]
+impl ScheduleProvider for Wrapper {
+    async fn start_auto_update_task(
+        &self,
+        cancellation_token: CancellationToken,
+    ) -> Result<(), Box<dyn std::error::Error + Send + Sync + 'static>> {
+        let mut ticker = interval(Duration::from_secs(60 * 30));
+        ticker.tick().await; // bc we already have the latest schedule, when instantiating provider
+
+        loop {
+            tokio::select! {
+                _ = ticker.tick() => {
+                    let mut lock = self.inner.write().await;
+                    let this= lock.deref_mut();
+
+                    log::info!("Updating schedule...");
+
+                    match this.updater.update(&mut this.snapshot).await {
+                        Ok(snapshot) => {
+                            this.snapshot = Arc::new(snapshot);
+                        },
+
+                        Err(err) => {
+                            cancellation_token.cancel();
+                            return Err(err.into());
+                        }
+                    }
+                }
+
+                _ = cancellation_token.cancelled() => {
+                    return Ok(());
+                }
+            }
+        }
+    }
+
+    async fn get_schedule(&self) -> Arc<ScheduleSnapshot> {
+        self.inner.read().await.snapshot.clone()
+    }
+}

providers/provider-engels-polytechnic/src/parser/macros.rs

@@ -0,0 +1,21 @@
+#[macro_export]
+macro_rules! or_continue {
+    ( $e:expr ) => {{
+        if let Some(x) = $e {
+            x
+        } else {
+            continue;
+        }
+    }};
+}
+
+#[macro_export]
+macro_rules! or_break {
+    ( $e:expr ) => {{
+        if let Some(x) = $e {
+            x
+        } else {
+            break;
+        }
+    }};
+}

providers/provider-engels-polytechnic/src/parser/mod.rs

@@ -0,0 +1,912 @@
+use crate::or_continue;
+use crate::parser::error::{ErrorCell, ErrorCellPos};
+use crate::parser::worksheet::WorkSheet;
+use crate::parser::LessonParseResult::{Lessons, Street};
+use base::LessonType::Break;
+use base::{
+    Day, Lesson, LessonBoundaries, LessonSubGroup, LessonType, ParsedSchedule, ScheduleEntry,
+};
+use calamine::{open_workbook_from_rs, Reader, Xls};
+use chrono::{DateTime, Duration, NaiveDate, NaiveTime, Utc};
+use regex::Regex;
+use std::collections::HashMap;
+use std::io::Cursor;
+use std::sync::LazyLock;
+
+mod macros;
+mod worksheet;
+
+pub mod error {
+    use derive_more::{Display, Error};
+    use serde::{Serialize, Serializer};
+    use std::sync::Arc;
+    use utoipa::ToSchema;
+
+    #[derive(Clone, Debug, Display, Error, ToSchema)]
+    #[display("row {row}, column {column}")]
+    pub struct ErrorCellPos {
+        pub row: u32,
+        pub column: u32,
+    }
+
+    #[derive(Clone, Debug, Display, Error, ToSchema)]
+    #[display("'{data}' at {pos}")]
+    pub struct ErrorCell {
+        pub pos: ErrorCellPos,
+        pub data: String,
+    }
+
+    impl ErrorCell {
+        pub fn new(row: u32, column: u32, data: String) -> Self {
+            Self {
+                pos: ErrorCellPos { row, column },
+                data,
+            }
+        }
+    }
+
+    #[derive(Clone, Debug, Display, Error, ToSchema)]
+    pub enum Error {
+        /// Errors related to reading XLS file.
+        #[display("{_0:?}: Failed to read XLS file.")]
+        #[schema(value_type = String)]
+        BadXLS(Arc<calamine::XlsError>),
+
+        /// Not a single sheet was found.
+        #[display("No work sheets found.")]
+        NoWorkSheets,
+
+        /// There are no data on the boundaries of the sheet.
+        #[display("There is no data on work sheet boundaries.")]
+        UnknownWorkSheetRange,
+
+        /// Failed to read the beginning and end of the lesson from the cell
+        #[display("Failed to read lesson start and end from {_0}.")]
+        LessonBoundaries(ErrorCell),
+
+        /// Not found the beginning and the end corresponding to the lesson.
+        #[display("No start and end times matching the lesson (at {_0}) was found.")]
+        LessonTimeNotFound(ErrorCellPos),
+    }
+
+    impl Serialize for Error {
+        fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error>
+        where
+            S: Serializer,
+        {
+            match self {
+                Error::BadXLS(_) => serializer.serialize_str("BAD_XLS"),
+                Error::NoWorkSheets => serializer.serialize_str("NO_WORK_SHEETS"),
+                Error::UnknownWorkSheetRange => {
+                    serializer.serialize_str("UNKNOWN_WORK_SHEET_RANGE")
+                }
+                Error::LessonBoundaries(_) => serializer.serialize_str("GLOBAL_TIME"),
+                Error::LessonTimeNotFound(_) => serializer.serialize_str("LESSON_TIME_NOT_FOUND"),
+            }
+        }
+    }
+}
+
+/// Data cell storing the group name.
+pub struct GroupCellInfo {
+    /// Column index.
+    pub column: u32,
+
+    /// Text in the cell.
+    pub name: String,
+}
+
+/// Data cell storing the line.
+pub struct DayCellInfo {
+    /// Line index.
+    pub row: u32,
+
+    /// Column index.
+    pub column: u32,
+
+    /// Day name.
+    pub name: String,
+
+    /// Date of the day.
+    pub date: DateTime<Utc>,
+}
+
+/// Data on the time of lessons from the second column of the schedule.
+pub struct BoundariesCellInfo {
+    /// Temporary segment of the lesson.
+    pub time_range: LessonBoundaries,
+
+    /// Type of lesson.
+    pub lesson_type: LessonType,
+
+    /// The lesson index.
+    pub default_index: Option<u32>,
+
+    /// The frame of the cell.
+    pub xls_range: ((u32, u32), (u32, u32)),
+}
+/// Obtaining a "skeleton" schedule from the working sheet.
+fn parse_skeleton(
+    worksheet: &WorkSheet,
+) -> Result<(Vec<DayCellInfo>, Vec<GroupCellInfo>), crate::parser::error::Error> {
+    let mut groups: Vec<GroupCellInfo> = Vec::new();
+    let mut days: Vec<(u32, String, Option<DateTime<Utc>>)> = Vec::new();
+
+    let worksheet_start = worksheet
+        .start()
+        .ok_or(error::Error::UnknownWorkSheetRange)?;
+    let worksheet_end = worksheet.end().ok_or(error::Error::UnknownWorkSheetRange)?;
+
+    let mut row = worksheet_start.0;
+
+    while row < worksheet_end.0 {
+        row += 1;
+
+        let day_full_name = or_continue!(worksheet.get_string_from_cell(row, 0));
+
+        // parse groups row when days column will found
+        if groups.is_empty() {
+            // переход на предыдущую строку
+            row -= 1;
+
+            for column in (worksheet_start.1 + 2)..=worksheet_end.1 {
+                groups.push(GroupCellInfo {
+                    column,
+                    name: or_continue!(worksheet.get_string_from_cell(row, column))
+                        .replace(" ", ""),
+                });
+            }
+
+            // возврат на текущую строку
+            row += 1;
+        }
+
+        let (day_name, day_date) = {
+            let space_index = match day_full_name.find(' ') {
+                Some(index) => {
+                    if index < 10 {
+                        break;
+                    } else {
+                        index
+                    }
+                }
+                None => break,
+            };
+
+            let name = day_full_name[..space_index].to_string();
+
+            let date_slice = &day_full_name[space_index + 1..];
+            let date = NaiveDate::parse_from_str(date_slice, "%d.%m.%Y")
+                .map(|date| date.and_time(NaiveTime::default()).and_utc())
+                .ok();
+
+            (name, date)
+        };
+
+        days.push((row, day_name, day_date));
+    }
+
+    // fix unparsable day dates
+    let days_max = days.len().min(5);
+
+    for i in 0..days_max {
+        if days[i].2.is_none() && days[i + 1].2.is_some() {
+            days[i].2 = Some(days[i + 1].2.unwrap() - Duration::days(1));
+        }
+    }
+
+    for i in 0..days_max {
+        let i = days_max - i;
+
+        if days[i - 1].2.is_none() && days[i].2.is_some() {
+            days[i - 1].2 = Some(days[i].2.unwrap() - Duration::days(1));
+        }
+    }
+
+    let days = days
+        .into_iter()
+        .map(|day| DayCellInfo {
+            row: day.0,
+            column: 0,
+            name: day.1,
+            date: day.2.unwrap(),
+        })
+        .collect();
+
+    Ok((days, groups))
+}
+
+/// The result of obtaining a lesson from the cell.
+enum LessonParseResult {
+    /// List of lessons long from one to two.
+    ///
+    /// The number of lessons will be equal to one if the couple is the first in the day,
+    /// otherwise the list from the change template and the lesson itself will be returned.
+    Lessons(Vec<Lesson>),
+
+    /// Street on which the Polytechnic Corps is located.
+    Street(String),
+}
+
+// noinspection GrazieInspection
+/// Obtaining a non-standard type of lesson by name.
+fn guess_lesson_type(text: &String) -> Option<LessonType> {
+    static MAP: LazyLock<HashMap<&str, LessonType>> = LazyLock::new(|| {
+        HashMap::from([
+            ("консультация", LessonType::Consultation),
+            ("самостоятельная работа", LessonType::IndependentWork),
+            ("зачет", LessonType::Exam),
+            ("зачет с оценкой", LessonType::ExamWithGrade),
+            ("экзамен", LessonType::ExamDefault),
+            ("курсовой проект", LessonType::CourseProject),
+            ("защита курсового проекта", LessonType::CourseProjectDefense),
+        ])
+    });
+
+    let name_lower = text.to_lowercase();
+
+    match MAP
+        .iter()
+        .map(|(text, lesson_type)| (lesson_type, strsim::levenshtein(text, &*name_lower)))
+        .filter(|x| x.1 <= 4)
+        .min_by_key(|(_, score)| *score)
+    {
+        None => None,
+        Some(v) => Some(v.0.clone()),
+    }
+}
+
+/// Getting a pair or street from a cell.
+fn parse_lesson(
+    worksheet: &WorkSheet,
+    day: &Day,
+    day_boundaries: &Vec<BoundariesCellInfo>,
+    lesson_boundaries: &BoundariesCellInfo,
+    group_column: u32,
+) -> Result<LessonParseResult, crate::parser::error::Error> {
+    let row = lesson_boundaries.xls_range.0.0;
+
+    let name = {
+        let cell_data = match worksheet.get_string_from_cell(row, group_column) {
+            Some(x) => x,
+            None => return Ok(Lessons(Vec::new())),
+        };
+
+        static OTHER_STREET_RE: LazyLock<Regex> =
+            LazyLock::new(|| Regex::new(r"^[А-Я][а-я]+[,\s]\d+$").unwrap());
+
+        if OTHER_STREET_RE.is_match(&cell_data) {
+            return Ok(Street(cell_data));
+        }
+
+        cell_data
+    };
+
+    let cell_range = worksheet.get_merge_from_start(row, group_column);
+
+    let (default_range, lesson_time) = {
+        let end_time_arr = day_boundaries
+            .iter()
+            .filter(|time| time.xls_range.1.0 == cell_range.1.0)
+            .collect::<Vec<&BoundariesCellInfo>>();
+
+        let end_time = end_time_arr
+            .first()
+            .ok_or(error::Error::LessonTimeNotFound(ErrorCellPos {
+                row,
+                column: group_column,
+            }))?;
+
+        let range: Option<[u8; 2]> = if lesson_boundaries.default_index != None {
+            let default = lesson_boundaries.default_index.unwrap() as u8;
+            Some([default, end_time.default_index.unwrap() as u8])
+        } else {
+            None
+        };
+
+        let time = LessonBoundaries {
+            start: lesson_boundaries.time_range.start,
+            end: end_time.time_range.end,
+        };
+
+        Ok((range, time))
+    }?;
+
+    let (name, mut subgroups, lesson_type) = parse_name_and_subgroups(&name)?;
+
+    {
+        let cabinets: Vec<String> = parse_cabinets(
+            worksheet,
+            (cell_range.0.0, cell_range.1.0),
+            group_column + 1,
+        );
+
+        let cab_count = cabinets.len();
+
+        if cab_count == 1 {
+            // Назначаем этот кабинет всем подгруппам
+            let cab = Some(cabinets.get(0).unwrap().clone());
+
+            for subgroup in &mut subgroups {
+                if let Some(subgroup) = subgroup {
+                    subgroup.cabinet = cab.clone()
+                }
+            }
+        } else if cab_count == 2 {
+            while subgroups.len() < cab_count {
+                subgroups.push(subgroups.last().unwrap_or(&None).clone());
+            }
+
+            for i in 0..cab_count {
+                let subgroup = subgroups.get_mut(i).unwrap();
+                let cabinet = Some(cabinets.get(i).unwrap().clone());
+
+                match subgroup {
+                    None => {
+                        let _ = subgroup.insert(LessonSubGroup {
+                            teacher: None,
+                            cabinet,
+                        });
+                    }
+                    Some(subgroup) => {
+                        subgroup.cabinet = cabinet;
+                    }
+                }
+            }
+        }
+    };
+
+    let lesson = Lesson {
+        lesson_type: lesson_type.unwrap_or(lesson_boundaries.lesson_type.clone()),
+        range: default_range,
+        name: Some(name),
+        time: lesson_time,
+        subgroups: if subgroups.len() == 2
+            && subgroups.get(0).unwrap().is_none()
+            && subgroups.get(1).unwrap().is_none()
+        {
+            None
+        } else {
+            Some(subgroups)
+        },
+        group: None,
+    };
+
+    let prev_lesson = if day.lessons.is_empty() {
+        return Ok(Lessons(Vec::from([lesson])));
+    } else {
+        &day.lessons[day.lessons.len() - 1]
+    };
+
+    Ok(Lessons(Vec::from([
+        Lesson {
+            lesson_type: Break,
+            range: None,
+            name: None,
+            time: LessonBoundaries {
+                start: prev_lesson.time.end,
+                end: lesson.time.start,
+            },
+            subgroups: Some(Vec::new()),
+            group: None,
+        },
+        lesson,
+    ])))
+}
+
+/// Obtaining a list of cabinets to the right of the lesson cell.
+fn parse_cabinets(worksheet: &WorkSheet, row_range: (u32, u32), column: u32) -> Vec<String> {
+    let mut cabinets: Vec<String> = Vec::new();
+
+    for row in row_range.0..row_range.1 {
+        let raw = or_continue!(worksheet.get_string_from_cell(row, column));
+
+        let clean = raw.replace("\n", " ");
+        let parts: Vec<&str> = clean.split(" ").collect();
+
+        parts.iter().take(2).for_each(|part| {
+            let clean_part = part.to_string().trim().to_string();
+
+            cabinets.push(clean_part);
+        });
+
+        break;
+    }
+
+    cabinets
+}
+
+//noinspection GrazieInspection
+/// Getting the "pure" name of the lesson and list of teachers from the text of the lesson cell.
+fn parse_name_and_subgroups(
+    text: &String,
+) -> Result<(String, Vec<Option<LessonSubGroup>>, Option<LessonType>), crate::parser::error::Error>
+{
+    // Части названия пары:
+    // 1. Само название.
+    // 2. Список преподавателей и подгрупп.
+    // 3. "Модификатор" (чаще всего).
+    //
+    // Регулярное выражение для получения ФИО преподавателей и номеров подгрупп (aka. второй части).
+    // (?:[А-Я][а-я]+\s?(?:[А-Я][\s.]*){2}(?:\(\d\s?[а-я]+\))?(?:, )?)+[\s.]*
+    //
+    // Подробнее:
+    // (?:
+    //     [А-Я][а-я]+         - Фамилия.
+    //     \s?                 - Кто знает, будет ли там пробел.
+    //     (?:[А-Я][\s.]*){2}  - Имя и отчество с учётом случайных пробелов и точек.
+    //     (?:
+    //         \(              - Открытие подгруппы.
+    //         \s?             - Кто знает, будет ли там пробел.
+    //         \d              - Номер подгруппы.
+    //         \s?             - Кто знает, будет ли там пробел.
+    //         [а-я\s]+        - Слово "подгруппа" с учётов ошибок.
+    //         \)              - Закрытие подгруппы.
+    //     )?                  - Явное указание подгруппы может отсутствовать по понятным причинам.
+    //     (?:, )?             - Разделители между отдельными частями.
+    // )+
+    // [\s.]*                  - Забираем с собой всякий мусор, что бы не передать его в третью часть.
+
+    static NAMES_REGEX: LazyLock<Regex> = LazyLock::new(|| {
+        Regex::new(
+            r"(?:[А-Я][а-я]+\s?(?:[А-Я][\s.]*){2}(?:\(\s*\d\s*[а-я\s]+\))?(?:[\s,]+)?){1,2}+[\s.,]*",
+        )
+            .unwrap()
+    });
+
+    // Отчистка
+    static CLEAN_RE: LazyLock<Regex> = LazyLock::new(|| Regex::new(r"[\s\n\t]+").unwrap());
+
+    let text = CLEAN_RE
+        .replace(&text.replace(&[' ', '\t', '\n'], " "), " ")
+        .to_string();
+
+    let (lesson_name, subgroups, lesson_type) = match NAMES_REGEX.captures(&text) {
+        Some(captures) => {
+            let capture = captures.get(0).unwrap();
+
+            let subgroups: Vec<Option<LessonSubGroup>> = {
+                let src = capture.as_str().replace(&[' ', '.'], "");
+
+                let mut shared_subgroup = false;
+                let mut subgroups: [Option<LessonSubGroup>; 2] = [None, None];
+
+                for name in src.split(',') {
+                    let open_bracket_index = name.find('(');
+
+                    let number: u8 = open_bracket_index
+                        .map_or(0, |index| name[(index + 1)..(index + 2)].parse().unwrap());
+
+                    let teacher_name = {
+                        let name_end = open_bracket_index.unwrap_or_else(|| name.len());
+
+                        // Я ебал. Как же я долго до этого доходил.
+                        format!(
+                            "{} {}.{}.",
+                            name.get(..name_end - 4).unwrap(),
+                            name.get(name_end - 4..name_end - 2).unwrap(),
+                            name.get(name_end - 2..name_end).unwrap(),
+                        )
+                    };
+
+                    let lesson = Some(LessonSubGroup {
+                        cabinet: None,
+                        teacher: Some(teacher_name),
+                    });
+
+                    match number {
+                        0 => {
+                            subgroups[0] = lesson;
+                            subgroups[1] = None;
+                            shared_subgroup = true;
+                            break;
+                        }
+                        num => {
+                            // 1 - 1 = 0 | 2 - 1 = 1 | 3 - 1 = 2 (schedule index to array index)
+                            // 0 % 2 = 0 | 1 % 2 = 1 | 2 % 2 = 0 (clamp)
+                            let normalised = (num - 1) % 2;
+
+                            subgroups[normalised as usize] = lesson;
+                        }
+                    }
+                }
+
+                if shared_subgroup {
+                    Vec::from([subgroups[0].take()])
+                } else {
+                    Vec::from(subgroups)
+                }
+            };
+
+            let name = text[..capture.start()].trim().to_string();
+            let extra = text[capture.end()..].trim().to_string();
+
+            let lesson_type = if extra.len() > 4 {
+                let result = guess_lesson_type(&extra);
+
+                if result.is_none() {
+                    #[cfg(not(debug_assertions))]
+                    sentry::capture_message(
+                        &*format!("Не удалось угадать тип пары '{}'!", extra),
+                        sentry::Level::Warning,
+                    );
+
+                    #[cfg(debug_assertions)]
+                    log::warn!("Не удалось угадать тип пары '{}'!", extra);
+                }
+
+                result
+            } else {
+                None
+            };
+
+            (name, subgroups, lesson_type)
+        }
+        None => (text, Vec::new(), None),
+    };
+
+    Ok((lesson_name, subgroups, lesson_type))
+}
+
+/// Getting the start and end of a pair from a cell in the first column of a document.
+///
+/// # Arguments
+///
+/// * `cell_data`: text in cell.
+/// * `date`: date of the current day.
+fn parse_lesson_boundaries_cell(
+    cell_data: &String,
+    date: DateTime<Utc>,
+) -> Option<LessonBoundaries> {
+    static TIME_RE: LazyLock<Regex> =
+        LazyLock::new(|| Regex::new(r"(\d+\.\d+)-(\d+\.\d+)").unwrap());
+
+    let parse_res = if let Some(captures) = TIME_RE.captures(cell_data) {
+        captures
+    } else {
+        return None;
+    };
+
+    let start_match = parse_res.get(1).unwrap().as_str();
+    let start_parts: Vec<&str> = start_match.split(".").collect();
+
+    let end_match = parse_res.get(2).unwrap().as_str();
+    let end_parts: Vec<&str> = end_match.split(".").collect();
+
+    static GET_TIME: fn(DateTime<Utc>, &Vec<&str>) -> DateTime<Utc> = |date, parts| {
+        date + Duration::hours(parts[0].parse::<i64>().unwrap() - 4)
+            + Duration::minutes(parts[1].parse::<i64>().unwrap())
+    };
+
+    Some(LessonBoundaries {
+        start: GET_TIME(date.clone(), &start_parts),
+        end: GET_TIME(date, &end_parts),
+    })
+}
+
+/// Parse the column of the document to obtain a list of day's lesson boundaries.
+///
+/// # Arguments
+///
+/// * `worksheet`: document.
+/// * `date`: date of the current day.
+/// * `row_range`: row boundaries of the current day.
+/// * `column`: column with the required data.
+fn parse_day_boundaries(
+    worksheet: &WorkSheet,
+    date: DateTime<Utc>,
+    row_range: (u32, u32),
+    column: u32,
+) -> Result<Vec<BoundariesCellInfo>, crate::parser::error::Error> {
+    let mut day_times: Vec<BoundariesCellInfo> = Vec::new();
+
+    for row in row_range.0..row_range.1 {
+        let time_cell = if let Some(str) = worksheet.get_string_from_cell(row, column) {
+            str
+        } else {
+            continue;
+        };
+
+        let lesson_time = parse_lesson_boundaries_cell(&time_cell, date.clone()).ok_or(
+            error::Error::LessonBoundaries(ErrorCell::new(row, column, time_cell.clone())),
+        )?;
+
+        // type
+        let lesson_type = if time_cell.contains("пара") {
+            LessonType::Default
+        } else {
+            LessonType::Additional
+        };
+
+        // lesson index
+        let default_index = if lesson_type == LessonType::Default {
+            Some(
+                time_cell
+                    .chars()
+                    .next()
+                    .unwrap()
+                    .to_string()
+                    .parse::<u32>()
+                    .unwrap(),
+            )
+        } else {
+            None
+        };
+
+        day_times.push(BoundariesCellInfo {
+            time_range: lesson_time,
+            lesson_type,
+            default_index,
+            xls_range: worksheet.get_merge_from_start(row, column),
+        });
+    }
+
+    Ok(day_times)
+}
+
+/// Parse the column of the document to obtain a list of week's lesson boundaries.
+///
+/// # Arguments
+///
+/// * `worksheet`: document.
+/// * `week_markup`: markup of the current week.
+fn parse_week_boundaries(
+    worksheet: &WorkSheet,
+    week_markup: &Vec<DayCellInfo>,
+) -> Result<Vec<Vec<BoundariesCellInfo>>, crate::parser::error::Error> {
+    let mut result: Vec<Vec<BoundariesCellInfo>> = Vec::new();
+
+    let worksheet_end_row = worksheet.end().unwrap().0;
+    let lesson_time_column = week_markup[0].column + 1;
+
+    for day_index in 0..week_markup.len() {
+        let day_markup = &week_markup[day_index];
+
+        // Если текущий день не последнему, то индекс строки следующего дня.
+        // Если текущий день - последний, то индекс последней строки документа.
+        let end_row = if day_index != week_markup.len() - 1 {
+            week_markup[day_index + 1].row
+        } else {
+            worksheet_end_row
+        };
+
+        let day_boundaries = parse_day_boundaries(
+            &worksheet,
+            day_markup.date.clone(),
+            (day_markup.row, end_row),
+            lesson_time_column,
+        )?;
+
+        result.push(day_boundaries);
+    }
+
+    Ok(result)
+}
+
+/// Conversion of the list of couples of groups in the list of lessons of teachers.
+fn convert_groups_to_teachers(
+    groups: &HashMap<String, ScheduleEntry>,
+) -> HashMap<String, ScheduleEntry> {
+    let mut teachers: HashMap<String, ScheduleEntry> = HashMap::new();
+
+    let empty_days: Vec<Day> = groups
+        .values()
+        .next()
+        .unwrap()
+        .days
+        .iter()
+        .map(|day| Day {
+            name: day.name.clone(),
+            street: day.street.clone(),
+            date: day.date.clone(),
+            lessons: vec![],
+        })
+        .collect();
+
+    for group in groups.values() {
+        for (index, day) in group.days.iter().enumerate() {
+            for group_lesson in &day.lessons {
+                if group_lesson.lesson_type == Break {
+                    continue;
+                }
+
+                if group_lesson.subgroups.is_none() {
+                    continue;
+                }
+
+                let subgroups = group_lesson.subgroups.as_ref().unwrap();
+
+                for subgroup in subgroups {
+                    let teacher = match subgroup {
+                        None => continue,
+                        Some(subgroup) => match &subgroup.teacher {
+                            None => continue,
+                            Some(teacher) => teacher,
+                        },
+                    };
+
+                    if teacher == "Ошибка в расписании" {
+                        continue;
+                    }
+
+                    if !teachers.contains_key(teacher) {
+                        teachers.insert(
+                            teacher.clone(),
+                            ScheduleEntry {
+                                name: teacher.clone(),
+                                days: empty_days.to_vec(),
+                            },
+                        );
+                    }
+
+                    let teacher_day = teachers
+                        .get_mut(teacher)
+                        .unwrap()
+                        .days
+                        .get_mut(index)
+                        .unwrap();
+
+                    teacher_day.lessons.push({
+                        let mut lesson = group_lesson.clone();
+                        lesson.group = Some(group.name.clone());
+
+                        lesson
+                    });
+                }
+            }
+        }
+    }
+
+    teachers.iter_mut().for_each(|(_, teacher)| {
+        teacher.days.iter_mut().for_each(|day| {
+            day.lessons
+                .sort_by(|a, b| a.range.as_ref().unwrap()[1].cmp(&b.range.as_ref().unwrap()[1]))
+        })
+    });
+
+    teachers
+}
+
+/// Reading XLS Document from the buffer and converting it into the schedule ready to use.
+///
+/// # Arguments
+///
+/// * `buffer`: XLS data containing schedule.
+///
+/// returns: Result<ParseResult, crate::parser::error::Error>
+///
+/// # Examples
+///
+/// ```
+/// use schedule_parser::parse_xls;
+///
+/// let result = parse_xls(&include_bytes!("../../schedule.xls").to_vec());
+///
+/// assert!(result.is_ok(), "{}", result.err().unwrap());
+///
+/// assert_ne!(result.as_ref().unwrap().groups.len(), 0);
+/// assert_ne!(result.as_ref().unwrap().teachers.len(), 0);
+/// ```
+pub fn parse_xls(buffer: &Vec<u8>) -> Result<ParsedSchedule, crate::parser::error::Error> {
+    let cursor = Cursor::new(&buffer);
+    let mut workbook: Xls<_> =
+        open_workbook_from_rs(cursor).map_err(|e| error::Error::BadXLS(std::sync::Arc::new(e)))?;
+
+    let worksheet = {
+        let (worksheet_name, worksheet) = workbook
+            .worksheets()
+            .first()
+            .ok_or(error::Error::NoWorkSheets)?
+            .clone();
+
+        let worksheet_merges = workbook
+            .worksheet_merge_cells(&*worksheet_name)
+            .ok_or(error::Error::NoWorkSheets)?;
+
+        WorkSheet {
+            data: worksheet,
+            merges: worksheet_merges,
+        }
+    };
+
+    let (week_markup, groups_markup) = parse_skeleton(&worksheet)?;
+    let week_boundaries = parse_week_boundaries(&worksheet, &week_markup)?;
+
+    let mut groups: HashMap<String, ScheduleEntry> = HashMap::new();
+
+    for group_markup in groups_markup {
+        let mut group = ScheduleEntry {
+            name: group_markup.name,
+            days: Vec::new(),
+        };
+
+        for day_index in 0..(&week_markup).len() {
+            let day_markup = &week_markup[day_index];
+
+            let mut day = Day {
+                name: day_markup.name.clone(),
+                street: None,
+                date: day_markup.date,
+                lessons: Vec::new(),
+            };
+
+            let day_boundaries = &week_boundaries[day_index];
+
+            for lesson_boundaries in day_boundaries {
+                match &mut parse_lesson(
+                    &worksheet,
+                    &day,
+                    &day_boundaries,
+                    &lesson_boundaries,
+                    group_markup.column,
+                )? {
+                    Lessons(lesson) => day.lessons.append(lesson),
+                    Street(street) => day.street = Some(street.to_owned()),
+                }
+            }
+
+            group.days.push(day);
+        }
+
+        groups.insert(group.name.clone(), group);
+    }
+
+    Ok(ParsedSchedule {
+        teachers: convert_groups_to_teachers(&groups),
+        groups,
+    })
+}
+
+#[cfg(any(test, feature = "test"))]
+pub mod test_utils {
+    use super::*;
+    use base::ParsedSchedule;
+
+    pub fn test_result() -> Result<ParsedSchedule, crate::parser::error::Error> {
+        parse_xls(&include_bytes!("../../../../test-data/engels-polytechnic.xls").to_vec())
+    }
+}
+
+#[cfg(test)]
+pub mod tests {
+    #[test]
+    fn read() {
+        let result = super::test_utils::test_result();
+
+        assert!(result.is_ok(), "{}", result.err().unwrap());
+
+        assert_ne!(result.as_ref().unwrap().groups.len(), 0);
+        assert_ne!(result.as_ref().unwrap().teachers.len(), 0);
+    }
+
+    #[test]
+    fn test_split_lesson() {
+        let result = super::test_utils::test_result();
+        assert!(result.is_ok(), "{}", result.err().unwrap());
+
+        let result = result.unwrap();
+        assert!(result.groups.contains_key("ИС-214/23"));
+
+        let group = result.groups.get("ИС-214/23").unwrap();
+
+        let thursday = group.days.get(3).unwrap();
+        assert_eq!(thursday.lessons.len(), 1);
+
+        let lesson = &thursday.lessons[0];
+        assert_eq!(lesson.range.unwrap()[1], 3);
+        assert!(lesson.subgroups.is_some());
+
+        let subgroups = lesson.subgroups.as_ref().unwrap();
+        assert_eq!(subgroups.len(), 2);
+
+        assert_eq!(
+            subgroups[0].as_ref().unwrap().cabinet,
+            Some("44".to_string())
+        );
+
+        assert_eq!(
+            subgroups[1].as_ref().unwrap().cabinet,
+            Some("43".to_string())
+        );
+    }
+}

providers/provider-engels-polytechnic/src/parser/worksheet.rs

@@ -0,0 +1,58 @@
+use regex::Regex;
+use std::ops::Deref;
+use std::sync::LazyLock;
+
+/// XLS WorkSheet data.
+pub struct WorkSheet {
+    pub data: calamine::Range<calamine::Data>,
+    pub merges: Vec<calamine::Dimensions>,
+}
+
+impl Deref for WorkSheet {
+    type Target = calamine::Range<calamine::Data>;
+
+    fn deref(&self) -> &Self::Target {
+        &self.data
+    }
+}
+
+impl WorkSheet {
+    /// Getting a line from the required cell.
+    pub fn get_string_from_cell(&self, row: u32, col: u32) -> Option<String> {
+        let cell_data = if let Some(data) = self.get((row as usize, col as usize)) {
+            data.to_string()
+        } else {
+            return None;
+        };
+
+        if cell_data.trim().is_empty() {
+            return None;
+        }
+
+        static NL_RE: LazyLock<Regex> = LazyLock::new(|| Regex::new(r"[\n\r]+").unwrap());
+        static SP_RE: LazyLock<Regex> = LazyLock::new(|| Regex::new(r"\s+").unwrap());
+
+        let trimmed_data = SP_RE
+            .replace_all(&NL_RE.replace_all(&cell_data, " "), " ")
+            .trim()
+            .to_string();
+
+        if trimmed_data.is_empty() {
+            None
+        } else {
+            Some(trimmed_data)
+        }
+    }
+
+    /// Obtaining the boundaries of the cell along its upper left coordinate.
+    pub fn get_merge_from_start(&self, row: u32, column: u32) -> ((u32, u32), (u32, u32)) {
+        match self
+            .merges
+            .iter()
+            .find(|merge| merge.start.0 == row && merge.start.1 == column)
+        {
+            Some(merge) => (merge.start, (merge.end.0 + 1, merge.end.1 + 1)),
+            None => ((row, column), (row + 1, column + 1)),
+        }
+    }
+}

providers/provider-engels-polytechnic/src/updater.rs

@@ -0,0 +1,263 @@
+use crate::parser::parse_xls;
+use crate::updater::error::{Error, QueryUrlError, SnapshotCreationError};
+use crate::xls_downloader::{FetchError, XlsDownloader};
+use base::ScheduleSnapshot;
+
+pub enum UpdateSource {
+    Prepared(ScheduleSnapshot),
+
+    Url(String),
+
+    GrabFromSite {
+        yandex_api_key: String,
+        yandex_func_id: String,
+    },
+}
+
+pub struct Updater {
+    downloader: XlsDownloader,
+    update_source: UpdateSource,
+}
+
+pub mod error {
+    use crate::xls_downloader::FetchError;
+    use derive_more::{Display, Error};
+
+    #[derive(Debug, Display, Error)]
+    pub enum Error {
+        /// An error occurred while querying the Yandex Cloud API for a URL.
+        ///
+        /// This may result from network failures, invalid API credentials, or issues with the Yandex Cloud Function invocation.
+        /// See [`QueryUrlError`] for more details about specific causes.
+        QueryUrlFailed(QueryUrlError),
+
+        /// The schedule snapshot creation process failed.
+        ///
+        /// This can happen due to URL conflicts (same URL already in use), failed network requests,
+        /// download errors, or invalid XLS file content. See [`SnapshotCreationError`] for details.
+        SnapshotCreationFailed(SnapshotCreationError),
+    }
+    /// Errors that may occur when querying the Yandex Cloud API to retrieve a URL.
+    #[derive(Debug, Display, Error)]
+    pub enum QueryUrlError {
+        /// Occurs when the request to the Yandex Cloud API fails.
+        ///
+        /// This may be due to network issues, invalid API key, incorrect function ID, or other
+        /// problems with the Yandex Cloud Function invocation.
+        #[display("An error occurred during the request to the Yandex Cloud API: {_0}")]
+        RequestFailed(reqwest::Error),
+    }
+
+    /// Errors that may occur during the creation of a schedule snapshot.
+    #[derive(Debug, Display, Error)]
+    pub enum SnapshotCreationError {
+        /// The URL is the same as the one already being used (no update needed).
+        #[display("The URL is the same as the one already being used.")]
+        SameUrl,
+
+        /// The URL query for the XLS file failed to execute, either due to network issues or invalid API parameters.
+        #[display("Failed to fetch URL: {_0}")]
+        FetchFailed(FetchError),
+
+        /// Downloading the XLS file content failed after successfully obtaining the URL.
+        #[display("Download failed: {_0}")]
+        DownloadFailed(FetchError),
+
+        /// The XLS file could not be parsed into a valid schedule format.
+        #[display("Schedule data is invalid: {_0}")]
+        InvalidSchedule(crate::parser::error::Error),
+    }
+}
+
+impl Updater {
+    /// Constructs a new `ScheduleSnapshot` by downloading and parsing schedule data from the specified URL.
+    ///
+    /// This method first checks if the provided URL is the same as the one already configured in the downloader.
+    /// If different, it updates the downloader's URL, fetches the XLS content, parses it, and creates a snapshot.
+    /// Errors are returned for URL conflicts, network issues, download failures, or invalid data.
+    ///
+    /// # Arguments
+    ///
+    /// * `downloader`: A mutable reference to an `XLSDownloader` implementation used to fetch and parse the schedule data.
+    /// * `url`: The source URL pointing to the XLS file containing schedule data.
+    ///
+    /// returns: Result<ScheduleSnapshot, SnapshotCreationError>
+    pub async fn new_snapshot(
+        downloader: &mut XlsDownloader,
+        url: String,
+    ) -> Result<ScheduleSnapshot, SnapshotCreationError> {
+        if downloader.url.as_ref().is_some_and(|_url| _url.eq(&url)) {
+            return Err(SnapshotCreationError::SameUrl);
+        }
+
+        let head_result = downloader.set_url(&*url).await.map_err(|error| {
+            if let FetchError::Unknown(error) = &error {
+                sentry::capture_error(&error);
+            }
+
+            SnapshotCreationError::FetchFailed(error)
+        })?;
+
+        let xls_data = downloader
+            .fetch(false)
+            .await
+            .map_err(|error| {
+                if let FetchError::Unknown(error) = &error {
+                    sentry::capture_error(&error);
+                }
+
+                SnapshotCreationError::DownloadFailed(error)
+            })?
+            .data
+            .unwrap();
+
+        let parse_result = parse_xls(&xls_data).map_err(|error| {
+            sentry::capture_error(&error);
+
+            SnapshotCreationError::InvalidSchedule(error)
+        })?;
+
+        Ok(ScheduleSnapshot {
+            fetched_at: head_result.requested_at,
+            updated_at: head_result.uploaded_at,
+            url,
+            data: parse_result,
+        })
+    }
+
+    /// Queries the Yandex Cloud Function (FaaS) to obtain a URL for the schedule file.
+    ///
+    /// This sends a POST request to the specified Yandex Cloud Function endpoint,
+    /// using the provided API key for authentication. The returned URI is combined
+    /// with the "https://politehnikum-eng.ru" base domain to form the complete URL.
+    ///
+    /// # Arguments
+    ///
+    /// * `api_key` - Authentication token for Yandex Cloud API
+    /// * `func_id` - ID of the target Yandex Cloud Function to invoke
+    ///
+    /// # Returns
+    ///
+    /// Result containing:
+    /// - `Ok(String)` - Complete URL constructed from the Function's response
+    /// - `Err(QueryUrlError)` - If the request or response processing fails
+    async fn query_url(api_key: &str, func_id: &str) -> Result<String, QueryUrlError> {
+        let client = reqwest::Client::new();
+
+        let uri = client
+            .post(format!(
+                "https://functions.yandexcloud.net/{}?integration=raw",
+                func_id
+            ))
+            .header("Authorization", format!("Api-Key {}", api_key))
+            .send()
+            .await
+            .map_err(|error| QueryUrlError::RequestFailed(error))?
+            .text()
+            .await
+            .map_err(|error| QueryUrlError::RequestFailed(error))?;
+
+        Ok(format!("https://politehnikum-eng.ru{}", uri.trim()))
+    }
+
+    /// Initializes the schedule by fetching the URL from the environment or Yandex Cloud Function (FaaS)
+    /// and creating a [`ScheduleSnapshot`] with the downloaded data.
+    ///
+    /// # Arguments
+    ///
+    /// * `downloader`: Mutable reference to an `XLSDownloader` implementation used to fetch and parse the schedule
+    /// * `app_env`: Reference to the application environment containing either a predefined URL or Yandex Cloud credentials
+    ///
+    /// # Returns
+    ///
+    /// Returns `Ok(())` if the snapshot was successfully initialized, or an `Error` if:
+    /// - URL query to Yandex Cloud failed ([`QueryUrlError`])
+    /// - Schedule snapshot creation failed ([`SnapshotCreationError`])
+    pub async fn new(update_source: UpdateSource) -> Result<(Self, ScheduleSnapshot), Error> {
+        let mut this = Updater {
+            downloader: XlsDownloader::new(),
+            update_source,
+        };
+
+        if let UpdateSource::Prepared(snapshot) = &this.update_source {
+            let snapshot = snapshot.clone();
+            return Ok((this, snapshot));
+        }
+
+        let url = match &this.update_source {
+            UpdateSource::Url(url) => {
+                log::info!("The default link {} will be used", url);
+                url.clone()
+            }
+            UpdateSource::GrabFromSite {
+                yandex_api_key,
+                yandex_func_id,
+            } => {
+                log::info!("Obtaining a link using FaaS...");
+                Self::query_url(yandex_api_key, yandex_func_id)
+                    .await
+                    .map_err(|error| Error::QueryUrlFailed(error))?
+            }
+            _ => unreachable!(),
+        };
+
+        log::info!("For the initial setup, a link {} will be used", url);
+
+        let snapshot = Self::new_snapshot(&mut this.downloader, url)
+            .await
+            .map_err(|error| Error::SnapshotCreationFailed(error))?;
+
+        log::info!("Schedule snapshot successfully created!");
+
+        Ok((this, snapshot))
+    }
+
+    /// Updates the schedule snapshot by querying the latest URL from FaaS and checking for changes.
+    /// If the URL hasn't changed, only updates the [`fetched_at`] timestamp. If changed, downloads
+    /// and parses the new schedule data.
+    ///
+    /// # Arguments
+    ///
+    /// * `downloader`: XLS file downloader used to fetch and parse the schedule data
+    /// * `app_env`: Application environment containing Yandex Cloud configuration and auto-update settings
+    ///
+    /// returns: `Result<(), Error>` - Returns error if URL query fails or schedule parsing encounters issues
+    ///
+    /// # Safety
+    ///
+    /// Use `unsafe` to access the initialized snapshot, guaranteed valid by prior `init()` call
+    pub async fn update(
+        &mut self,
+        current_snapshot: &ScheduleSnapshot,
+    ) -> Result<ScheduleSnapshot, Error> {
+        if let UpdateSource::Prepared(snapshot) = &self.update_source {
+            let mut snapshot = snapshot.clone();
+            snapshot.update();
+            return Ok(snapshot);
+        }
+
+        let url = match &self.update_source {
+            UpdateSource::Url(url) => url.clone(),
+            UpdateSource::GrabFromSite {
+                yandex_api_key,
+                yandex_func_id,
+            } => Self::query_url(yandex_api_key.as_str(), yandex_func_id.as_str())
+                .await
+                .map_err(|error| Error::QueryUrlFailed(error))?,
+            _ => unreachable!(),
+        };
+
+        let snapshot = match Self::new_snapshot(&mut self.downloader, url).await {
+            Ok(snapshot) => snapshot,
+            Err(SnapshotCreationError::SameUrl) => {
+                let mut clone = current_snapshot.clone();
+                clone.update();
+
+                clone
+            }
+            Err(error) => return Err(Error::SnapshotCreationFailed(error)),
+        };
+
+        Ok(snapshot)
+    }
+}

providers/provider-engels-polytechnic/src/xls_downloader.rs

@@ -0,0 +1,237 @@
+use chrono::{DateTime, Utc};
+use derive_more::{Display, Error};
+use std::mem::discriminant;
+use std::sync::Arc;
+use utoipa::ToSchema;
+
+/// XLS data retrieval errors.
+#[derive(Clone, Debug, ToSchema, Display, Error)]
+pub enum FetchError {
+    /// File url is not set.
+    #[display("The link to the timetable was not provided earlier.")]
+    NoUrlProvided,
+
+    /// Unknown error.
+    #[display("An unknown error occurred while downloading the file.")]
+    #[schema(value_type = String)]
+    Unknown(Arc<reqwest::Error>),
+
+    /// Server returned a status code different from 200.
+    #[display("Server returned a status code {status_code}.")]
+    BadStatusCode { status_code: u16 },
+
+    /// The url leads to a file of a different type.
+    #[display("The link leads to a file of type '{content_type}'.")]
+    BadContentType { content_type: String },
+
+    /// Server doesn't return expected headers.
+    #[display("Server doesn't return expected header(s) '{expected_header}'.")]
+    BadHeaders { expected_header: String },
+}
+
+impl FetchError {
+    pub fn unknown(error: Arc<reqwest::Error>) -> Self {
+        Self::Unknown(error)
+    }
+
+    pub fn bad_status_code(status_code: u16) -> Self {
+        Self::BadStatusCode { status_code }
+    }
+
+    pub fn bad_content_type(content_type: &str) -> Self {
+        Self::BadContentType {
+            content_type: content_type.to_string(),
+        }
+    }
+
+    pub fn bad_headers(expected_header: &str) -> Self {
+        Self::BadHeaders {
+            expected_header: expected_header.to_string(),
+        }
+    }
+}
+
+impl PartialEq for FetchError {
+    fn eq(&self, other: &Self) -> bool {
+        discriminant(self) == discriminant(other)
+    }
+}
+
+/// Result of XLS data retrieval.
+#[derive(Debug, PartialEq)]
+pub struct FetchOk {
+    /// File upload date.
+    pub uploaded_at: DateTime<Utc>,
+
+    /// Date data received.
+    pub requested_at: DateTime<Utc>,
+
+    /// File data.
+    pub data: Option<Vec<u8>>,
+}
+
+impl FetchOk {
+    /// Result without file content.
+    pub fn head(uploaded_at: DateTime<Utc>) -> Self {
+        FetchOk {
+            uploaded_at,
+            requested_at: Utc::now(),
+            data: None,
+        }
+    }
+
+    /// Full result.
+    pub fn get(uploaded_at: DateTime<Utc>, data: Vec<u8>) -> Self {
+        FetchOk {
+            uploaded_at,
+            requested_at: Utc::now(),
+            data: Some(data),
+        }
+    }
+}
+
+pub type FetchResult = Result<FetchOk, FetchError>;
+
+pub struct XlsDownloader {
+    pub url: Option<String>,
+}
+
+impl XlsDownloader {
+    pub fn new() -> Self {
+        XlsDownloader { url: None }
+    }
+
+    async fn fetch_specified(url: &str, head: bool) -> FetchResult {
+        let client = reqwest::Client::new();
+
+        let response = if head {
+            client.head(url)
+        } else {
+            client.get(url)
+        }
+        .header("User-Agent", ua_generator::ua::spoof_chrome_ua())
+        .send()
+        .await
+        .map_err(|e| FetchError::unknown(Arc::new(e)))?;
+
+        if response.status().as_u16() != 200 {
+            return Err(FetchError::bad_status_code(response.status().as_u16()));
+        }
+
+        let headers = response.headers();
+
+        let content_type = headers
+            .get("Content-Type")
+            .ok_or(FetchError::bad_headers("Content-Type"))?;
+
+        if !headers.contains_key("etag") {
+            return Err(FetchError::bad_headers("etag"));
+        }
+
+        let last_modified = headers
+            .get("last-modified")
+            .ok_or(FetchError::bad_headers("last-modified"))?;
+
+        if content_type != "application/vnd.ms-excel" {
+            return Err(FetchError::bad_content_type(content_type.to_str().unwrap()));
+        }
+
+        let last_modified = DateTime::parse_from_rfc2822(&last_modified.to_str().unwrap())
+            .unwrap()
+            .with_timezone(&Utc);
+
+        Ok(if head {
+            FetchOk::head(last_modified)
+        } else {
+            FetchOk::get(last_modified, response.bytes().await.unwrap().to_vec())
+        })
+    }
+
+    pub async fn fetch(&self, head: bool) -> FetchResult {
+        if self.url.is_none() {
+            Err(FetchError::NoUrlProvided)
+        } else {
+            Self::fetch_specified(&*self.url.as_ref().unwrap(), head).await
+        }
+    }
+
+    pub async fn set_url(&mut self, url: &str) -> FetchResult {
+        let result = Self::fetch_specified(url, true).await;
+
+        if let Ok(_) = result {
+            self.url = Some(url.to_string());
+        }
+
+        result
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use crate::xls_downloader::{FetchError, XlsDownloader};
+
+    #[tokio::test]
+    async fn bad_url() {
+        let url = "bad_url";
+
+        let mut downloader = XlsDownloader::new();
+        assert!(downloader.set_url(url).await.is_err());
+    }
+
+    #[tokio::test]
+    async fn bad_status_code() {
+        let url = "https://www.google.com/not-found";
+
+        let mut downloader = XlsDownloader::new();
+        assert_eq!(
+            downloader.set_url(url).await,
+            Err(FetchError::bad_status_code(404))
+        );
+    }
+
+    #[tokio::test]
+    async fn bad_headers() {
+        let url = "https://www.google.com/favicon.ico";
+
+        let mut downloader = XlsDownloader::new();
+        assert_eq!(
+            downloader.set_url(url).await,
+            Err(FetchError::BadHeaders {
+                expected_header: "ETag".to_string(),
+            })
+        );
+    }
+
+    #[tokio::test]
+    async fn bad_content_type() {
+        let url = "https://s3.aero-storage.ldragol.ru/679e5d1145a6ad00843ad3f1/67ddb59fd46303008396ac96%2Fexample.txt";
+
+        let mut downloader = XlsDownloader::new();
+        assert!(downloader.set_url(url).await.is_err());
+    }
+
+    #[tokio::test]
+    async fn ok() {
+        let url = "https://s3.aero-storage.ldragol.ru/679e5d1145a6ad00843ad3f1/67ddb5fad46303008396ac97%2Fschedule.xls";
+
+        let mut downloader = XlsDownloader::new();
+        assert!(downloader.set_url(url).await.is_ok());
+    }
+
+    #[tokio::test]
+    async fn downloader_ok() {
+        let url = "https://s3.aero-storage.ldragol.ru/679e5d1145a6ad00843ad3f1/67ddb5fad46303008396ac97%2Fschedule.xls";
+
+        let mut downloader = XlsDownloader::new();
+        assert!(downloader.set_url(url).await.is_ok());
+        assert!(downloader.fetch(false).await.is_ok());
+    }
+
+    #[tokio::test]
+    async fn downloader_no_url_provided() {
+        let downloader = XlsDownloader::new();
+
+        let result = downloader.fetch(false).await;
+        assert_eq!(result, Err(FetchError::NoUrlProvided));
+    }
+}