fix(parser): fix sentry error sending

This allows us to not worry about subgroups array index overflows, and we can make better non-standard case solving.

providers/base/src/lib.rs

@@ -100,9 +100,9 @@ pub enum LessonType {
 
     /// Защита курсового проекта.
     CourseProjectDefense,
-    
+
     /// Практическое занятие.
-    Practice
+    Practice,
 }
 
 #[derive(Clone, Hash, Debug, Serialize, Deserialize, ToSchema)]
@@ -212,70 +212,6 @@ impl ScheduleSnapshot {
     }
 }
 
-// #[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 ParseError {
-//     /// 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 ParseError {
-//     fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error>
-//     where
-//         S: Serializer,
-//     {
-//         match self {
-//             ParseError::BadXLS(_) => serializer.serialize_str("BAD_XLS"),
-//             ParseError::NoWorkSheets => serializer.serialize_str("NO_WORK_SHEETS"),
-//             ParseError::UnknownWorkSheetRange => {
-//                 serializer.serialize_str("UNKNOWN_WORK_SHEET_RANGE")
-//             }
-//             ParseError::LessonBoundaries(_) => serializer.serialize_str("GLOBAL_TIME"),
-//             ParseError::LessonTimeNotFound(_) => serializer.serialize_str("LESSON_TIME_NOT_FOUND"),
-//         }
-//     }
-// }
-
 #[async_trait]
 pub trait ScheduleProvider
 where

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

@@ -1,21 +1,5 @@
-use derive_more::{Display, Error, From};
 use crate::parser::worksheet::CellPos;
-
-#[derive(Clone, Debug, Display, Error)]
-#[display("'{data}' at {pos}")]
-pub struct ErrorCell {
-    pub pos: CellPos,
-    pub data: String,
-}
-
-impl ErrorCell {
-    pub fn new(row: u32, column: u32, data: &str) -> Self {
-        Self {
-            pos: CellPos { row, column },
-            data: data.to_string(),
-        }
-    }
-}
+use derive_more::{Display, Error, From};
 
 #[derive(Debug, Display, Error, From)]
 pub enum Error {
@@ -28,11 +12,14 @@ pub enum Error {
     #[display("There is no data on work sheet boundaries.")]
     UnknownWorkSheetRange,
 
-    #[display("Failed to read lesson start and end from {_0}.")]
-    NoLessonBoundaries(ErrorCell),
+    #[display("Failed to read lesson start and end of lesson at {_0}.")]
+    NoLessonBoundaries(CellPos),
 
     #[display("No start and end times matching the lesson (at {_0}) was found.")]
     LessonTimeNotFound(CellPos),
+
+    #[display("Unknown lesson type `{type}` at {pos}")]
+    UnknownLessonType { pos: CellPos, r#type: String },
 }
 
 pub type Result<T> = core::result::Result<T, Error>;

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

@@ -1,6 +1,5 @@
 pub use self::error::{Error, Result};
 use crate::or_continue;
-use crate::parser::error::ErrorCell;
 use crate::parser::worksheet::{CellPos, CellRange, WorkSheet};
 use crate::parser::LessonParseResult::{Lessons, Street};
 use base::LessonType::Break;
@@ -217,7 +216,7 @@ fn parse_lesson(
         };
 
         static OTHER_STREET_RE: LazyLock<Regex> =
-            LazyLock::new(|| Regex::new(r"^[А-Я][а-я]+[,\s]\d+$").unwrap());
+            LazyLock::new(|| Regex::new(r"^[А-Я][а-я]+[,\s]+д\.\s\d+$").unwrap());
 
         if OTHER_STREET_RE.is_match(&cell_data) {
             return Ok(Street(cell_data));
@@ -226,12 +225,17 @@ fn parse_lesson(
         cell_data
     };
 
-    let cell_range = worksheet.get_merge_from_start(row, group_column);
+    let lesson_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.range.end.row == cell_range.end.row)
+            .filter(
+                |BoundariesData {
+                     range: CellRange { end, .. },
+                     ..
+                 }| { lesson_cell_range.end.row <= end.row },
+            )
             .collect::<Vec<&BoundariesData>>();
 
         let end_time = end_time_arr
@@ -257,12 +261,12 @@ fn parse_lesson(
         name,
         mut subgroups,
         r#type: lesson_type,
-    } = parse_name_and_subgroups(&name)?;
+    } = parse_name_and_subgroups(&name, row, group_column)?;
 
     {
         let cabinets: Vec<String> = parse_cabinets(
             worksheet,
-            (cell_range.start.row, cell_range.end.row),
+            (lesson_cell_range.start.row, lesson_cell_range.end.row),
             group_column + 1,
         );
 
@@ -364,7 +368,7 @@ struct ParsedLessonName {
 
 //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: &str) -> Result<ParsedLessonName> {
+fn parse_name_and_subgroups(text: &str, row: u32, column: u32) -> Result<ParsedLessonName> {
     // Части названия пары:
     // 1. Само название.
     // 2. Список преподавателей и подгрупп.
@@ -373,7 +377,7 @@ fn parse_name_and_subgroups(text: &str) -> Result<ParsedLessonName> {
     // Регулярное выражение для получения ФИО преподавателей и номеров подгрупп (aka. второй части).
     static NAME_RE: LazyLock<fancy_regex::Regex> = LazyLock::new(|| {
         fancy_regex::Regex::new(
-            r"([А-Я][а-я]+(?:[\s.]*[А-Я]){1,2})(?=[^а-я])[.\s]*(?:\(?(\d)[\sа-я]*\)?)?",
+            r"([А-Я][а-я]+(?:[\s.]*[А-Я]){1,2})(?=[^А-Яа-я])[.\s]*(?:\(?(\d)[\sа-я]*\)?)?",
         )
         .unwrap()
     });
@@ -394,10 +398,10 @@ fn parse_name_and_subgroups(text: &str) -> Result<ParsedLessonName> {
     let mut lesson_name: Option<&str> = None;
     let mut extra: Option<&str> = None;
 
-    let mut shared_subgroup = false;
+    let mut shared_subgroup = true;
     let mut subgroups: [Option<LessonSubGroup>; 2] = [None, None];
 
-    for capture in NAME_RE.captures_iter(&text) {
+    for capture in NAME_RE.captures_iter(&text).take(2) {
         let capture = capture.unwrap();
 
         if lesson_name.is_none() {
@@ -438,17 +442,23 @@ fn parse_name_and_subgroups(text: &str) -> Result<ParsedLessonName> {
 
         match subgroup_index {
             None => {
-                subgroups[0] = subgroup;
-                subgroups[1] = None;
-                shared_subgroup = true;
-                break;
+                // we have only 2 matches max so more than 2 subgroups we cant have 100%
+                *subgroups.iter_mut().find(|x| x.is_none()).unwrap() = subgroup;
             }
             Some(num) => {
+                // bc we have indexed subgroup
+                shared_subgroup = false;
+
                 // 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;
+                let subgroup_index = ((num - 1) % 2) as usize;
 
-                subgroups[normalised as usize] = subgroup;
+                // if we have subgroup in that index (probably non-indexed, we change it index to free)
+                if subgroups[subgroup_index].is_some() {
+                    subgroups.swap(0, 1);
+                }
+
+                subgroups[subgroup_index] = subgroup;
             }
         }
     }
@@ -456,7 +466,7 @@ fn parse_name_and_subgroups(text: &str) -> Result<ParsedLessonName> {
     let subgroups = if lesson_name.is_none() {
         Vec::new()
     } else if shared_subgroup {
-        Vec::from([subgroups[0].take()])
+        Vec::from([subgroups.into_iter().next().unwrap()])
     } else {
         Vec::from(subgroups)
     };
@@ -475,13 +485,19 @@ fn parse_name_and_subgroups(text: &str) -> Result<ParsedLessonName> {
 
         if result.is_none() {
             #[cfg(not(debug_assertions))]
-            sentry::capture_message(
-                &format!("Не удалось угадать тип пары '{}'!", extra),
-                sentry::Level::Warning,
-            );
+            sentry::capture_error(&Error::UnknownLessonType {
+                r#type: extra.to_string(),
+                pos: CellPos::new(row, column),
+            });
 
             #[cfg(debug_assertions)]
-            log::warn!("Не удалось угадать тип пары '{}'!", extra);
+            log::warn!(
+                "{}",
+                Error::UnknownLessonType {
+                    r#type: extra.to_string(),
+                    pos: CellPos::new(row, column),
+                }
+            );
         }
 
         result
@@ -548,9 +564,8 @@ fn parse_day_boundaries(
             continue;
         };
 
-        let lesson_time = parse_lesson_boundaries_cell(&time_cell, date).ok_or(
-            Error::NoLessonBoundaries(ErrorCell::new(row, column, &time_cell)),
-        )?;
+        let lesson_time = parse_lesson_boundaries_cell(&time_cell, date)
+            .ok_or(Error::NoLessonBoundaries(CellPos::new(row, column)))?;
 
         // type
         let lesson_type = if time_cell.contains("пара") {

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

@@ -1,5 +1,5 @@
-use derive_more::Display;
 use regex::Regex;
+use std::fmt::{Display, Formatter};
 use std::ops::Deref;
 use std::sync::LazyLock;
 
@@ -9,13 +9,35 @@ pub struct WorkSheet {
     pub merges: Vec<calamine::Dimensions>,
 }
 
-#[derive(Clone, Debug, Display, derive_more::Error)]
-#[display("row {row}, column {column}")]
+#[derive(Clone, Debug, derive_more::Error)]
 pub struct CellPos {
     pub row: u32,
     pub column: u32,
 }
 
+fn format_column_index(index: u32) -> String {
+    // https://stackoverflow.com/a/297214
+    let quotient = index / 26;
+
+    let char = char::from((65 + (index % 26)) as u8);
+
+    if quotient > 0 {
+        return format!("{}{}", format_column_index(quotient - 1), char);
+    }
+
+    return char.to_string();
+}
+
+impl Display for CellPos {
+    fn fmt(&self, f: &mut Formatter<'_>) -> std::fmt::Result {
+        f.write_fmt(format_args!(
+            "column {}, row {}",
+            format_column_index(self.column),
+            self.row + 1,
+        ))
+    }
+}
+
 pub struct CellRange {
     pub start: CellPos,
     pub end: CellPos,