Don't count multiple entries in a single row in column counts

[tablify] / src / lib.rs

diff --git a/src/lib.rs b/src/lib.rs
index 0584cd20406662282a66cb7466bd371ab1b0b5a7..c7271f3932965c7a2f64712764307af83e1ec593 100644 (file)
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -1,4 +1,6 @@
 #[cfg(test)]
 #[cfg(test)]

+use std::collections::{HashMap, HashSet};
+#[cfg(test)]

 use std::io::BufRead;
 #[cfg(test)]
 use std::iter::Iterator;
 use std::io::BufRead;
 #[cfg(test)]
 use std::iter::Iterator;


@@ -9,17 +11,74 @@ struct RowInput {
     entries: Vec<String>,
 }
 
     entries: Vec<String>,
 }
 

+struct Reader<Input: Iterator<Item = Result<String, std::io::Error>>> {
+    input: std::iter::Enumerate<Input>,
+    row: Option<RowInput>,
+}
+impl<Input: Iterator<Item = Result<String, std::io::Error>>> Reader<Input> {
+    #[cfg(test)]
+    fn new(input: Input) -> Self {
+        Self {
+            input: input.enumerate(),
+            row: None,
+        }
+    }
+}
+impl<Input: Iterator<Item = Result<String, std::io::Error>>> Iterator for Reader<Input> {
+    type Item = Result<RowInput, std::io::Error>;
+    fn next(&mut self) -> Option<Self::Item> {
+        loop {
+            match self
+                .input
+                .next()
+                .map(|(n, r)| (n, r.map(|line| String::from(line.trim_end()))))
+            {
+                None => return Ok(std::mem::take(&mut self.row)).transpose(),
+                Some((_, Err(e))) => return Some(Err(e)),
+                Some((_, Ok(line))) if line.is_empty() && self.row.is_some() => {
+                    return Ok(std::mem::take(&mut self.row)).transpose()
+                }
+                Some((_, Ok(line))) if line.is_empty() => {}
+                Some((n, Ok(line))) if line.starts_with(' ') => match &mut self.row {
+                    None => {
+                        return Some(Err(std::io::Error::other(format!(
+                            "{}: Entry with no header",
+                            n + 1
+                        ))))
+                    }
+                    Some(ref mut row) => row.entries.push(String::from(line.trim())),
+                },
+                Some((_, Ok(line))) => {
+                    let prev = std::mem::take(&mut self.row);
+                    self.row = Some(RowInput {
+                        label: line,
+                        entries: vec![],
+                    });
+                    if prev.is_some() {
+                        return Ok(prev).transpose();
+                    }
+                }
+            }
+        }
+    }
+}
+

 #[cfg(test)]
 #[cfg(test)]

-fn read_rows(input: impl std::io::Read) -> impl Iterator<Item = RowInput> {
-    vec![RowInput {
-        label: std::io::BufReader::new(input)
-            .lines()
-            .nth(0)
-            .unwrap()
-            .unwrap(),
-        entries: vec![],
-    }]
-    .into_iter()
+fn read_rows(input: impl std::io::Read) -> impl Iterator<Item = Result<RowInput, std::io::Error>> {
+    Reader::new(std::io::BufReader::new(input).lines())
+}
+
+#[cfg(test)]
+fn column_counts(rows: &[RowInput]) -> HashMap<String, usize> {
+    rows.iter()
+        .flat_map(|r| r.entries.iter().collect::<HashSet<_>>().into_iter())
+        .fold(HashMap::new(), |mut counts, e| {
+            counts
+                .entry(String::from(e))
+                .and_modify(|c| *c += 1)
+                .or_insert(1);
+            counts
+        })

 }
 
 pub fn tablify(_input: &impl std::io::Read) -> String {
 }
 
 pub fn tablify(_input: &impl std::io::Read) -> String {


@@ -33,18 +92,122 @@ mod tests {
     #[test]
     fn test_read_rows() {
         assert_eq!(
     #[test]
     fn test_read_rows() {
         assert_eq!(

-            read_rows(&b"foo"[..]).collect::<Vec<_>>(),
+            read_rows(&b"foo"[..]).flatten().collect::<Vec<_>>(),

             vec![RowInput {
                 label: String::from("foo"),
                 entries: vec![]
             }]
         );
         assert_eq!(
             vec![RowInput {
                 label: String::from("foo"),
                 entries: vec![]
             }]
         );
         assert_eq!(

-            read_rows(&b"bar"[..]).collect::<Vec<_>>(),
+            read_rows(&b"bar"[..]).flatten().collect::<Vec<_>>(),

             vec![RowInput {
                 label: String::from("bar"),
                 entries: vec![]
             }]
         );
             vec![RowInput {
                 label: String::from("bar"),
                 entries: vec![]
             }]
         );

+        assert_eq!(
+            read_rows(&b"foo\nbar\n"[..]).flatten().collect::<Vec<_>>(),
+            vec![
+                RowInput {
+                    label: String::from("foo"),
+                    entries: vec![]
+                },
+                RowInput {
+                    label: String::from("bar"),
+                    entries: vec![]
+                }
+            ]
+        );
+        assert_eq!(
+            read_rows(&b"foo\n bar\n"[..]).flatten().collect::<Vec<_>>(),
+            vec![RowInput {
+                label: String::from("foo"),
+                entries: vec![String::from("bar")]
+            }]
+        );
+        assert_eq!(
+            read_rows(&b"foo\n bar\n baz\n"[..])
+                .flatten()
+                .collect::<Vec<_>>(),
+            vec![RowInput {
+                label: String::from("foo"),
+                entries: vec![String::from("bar"), String::from("baz")]
+            }]
+        );
+        assert_eq!(
+            read_rows(&b"foo\n\nbar\n"[..])
+                .flatten()
+                .collect::<Vec<_>>(),
+            vec![
+                RowInput {
+                    label: String::from("foo"),
+                    entries: vec![]
+                },
+                RowInput {
+                    label: String::from("bar"),
+                    entries: vec![]
+                }
+            ]
+        );
+        assert_eq!(
+            read_rows(&b"foo\n \nbar\n"[..])
+                .flatten()
+                .collect::<Vec<_>>(),
+            vec![
+                RowInput {
+                    label: String::from("foo"),
+                    entries: vec![]
+                },
+                RowInput {
+                    label: String::from("bar"),
+                    entries: vec![]
+                }
+            ]
+        );
+        assert_eq!(
+            read_rows(&b"foo  \n bar  \n"[..])
+                .flatten()
+                .collect::<Vec<_>>(),
+            vec![RowInput {
+                label: String::from("foo"),
+                entries: vec![String::from("bar")]
+            }]
+        );
+
+        let bad = read_rows(&b" foo"[..]).next().unwrap();
+        assert!(bad.is_err());
+        assert!(format!("{bad:?}").contains("1: Entry with no header"));
+
+        let bad2 = read_rows(&b"foo\n\n bar"[..]).nth(1).unwrap();
+        assert!(bad2.is_err());
+        assert!(format!("{bad2:?}").contains("3: Entry with no header"));
+    }
+
+    #[test]
+    fn test_column_counts() {
+        assert_eq!(
+            column_counts(
+                &read_rows(&b"foo\n bar\n baz\n"[..])
+                    .collect::<Result<Vec<_>, _>>()
+                    .unwrap()
+            ),
+            HashMap::from([(String::from("bar"), 1), (String::from("baz"), 1)])
+        );
+        assert_eq!(
+            column_counts(
+                &read_rows(&b"foo\n bar\n baz\nquux\n baz"[..])
+                    .collect::<Result<Vec<_>, _>>()
+                    .unwrap()
+            ),
+            HashMap::from([(String::from("bar"), 1), (String::from("baz"), 2)])
+        );
+        assert_eq!(
+            column_counts(
+                &read_rows(&b"foo\n bar\n bar\n baz\n bar\nquux\n baz"[..])
+                    .collect::<Result<Vec<_>, _>>()
+                    .unwrap()
+            ),
+            HashMap::from([(String::from("bar"), 1), (String::from("baz"), 2)])
+        );

     }
 }
     }
 }