Give an example of applying a different template

- infobox requirements
- wikidata entries

Cargo.toml

@@ -8,7 +8,7 @@ This analysis tool allows seeing the wikidata behind articles through the lens o
 
 authors = ["Raymon Zutekouw <contact@raymon.dev>"]
 
-repository = ""
+repository = "https://codeberg.org/MeltingSilicon/wikipedia-infobox-analyzer"
 license = "MIT OR Apache-2.0"
 
 keywords = ["wikipedia", "wikidata", "infobox"]

README.md

@@ -31,6 +31,55 @@ By default, the tool assumes that you are looking for tools on the English wikip
 Make sure the passed title matches the article title, and the tool should be able to find the wikidata entry.
 The next section will go over what these infobox templates files are, where you can find them on wikipedia, and how you can customize them locally for your wikidata analysis.
 
+### Interpretation of output
+The left column of the table first lists all the properties that the infobox requires.
+The right column will display the same property name if it is present in wikidata.
+If it is not present, the right column will be blank.
+Remaining properties in wikidata that are not required by the infobox are trailed in the right column.
+
+``` markdown
+┌──────────────────┬─────────────────────┐
+│ Infobox requires │ Wikidata Earth (Q2) │
+├──────────────────┼─────────────────────┤
+│ P18              │ P18                 │ // property: image
+│ P170             │ P170                │ // property: creator
+│ P571             │ P571                │ // property: inception
+│                  │ P31                 │ // property: instance of
+│                  │ P138                │ // property: inception
+│                  │ P361                │ // property: part of
+│                  │ ....                │
+└──────────────────┴─────────────────────┘
+```
+
+This (shortened) example is complete, but if we were to try a different infobox template on the earth entry you can see that it is not a good fit.
+To demonstrate, we can apply the software template to the earth entry:
+
+``` markdown
+┌──────────────────┬─────────────────────┐
+│ Infobox requires │ Wikidata Earth (Q2) │
+├──────────────────┼─────────────────────┤
+│ P18              │ P18                 │ // property: image
+│ P154             │                     │
+│ P170             │ P170                │ // property: creator
+│ P178             │                     │
+│ P275             │                     │
+│ P277             │                     │
+│ P306             │                     │
+│ P348             │                     │
+│ P400             │                     │
+│ P548             │                     │
+│ P571             │ P571                │ // property: inception
+│ P577             │                     │
+│ P856             │                     │
+│ P1324            │                     │
+│ P2096            │                     │
+│                  │ P10                 │
+│                  │ P31                 │
+│                  │ P138                │
+│                  │ ....                │
+└──────────────────┴─────────────────────┘
+```
+
 ## Templates
 The wikipedia sites vary a lot when it comes to templates across the different languages.
 The goal of this tool is to be universal, but these templates have not been standardized as far as I am aware.
@@ -45,7 +94,7 @@ Then you can add the following line to that file locally:
 
 The program ignores what you put at the `...`.
 It permits templates that include a listing of wikidata entries for their templates.
-As an example, this is the case on the following template on the Dutch wikipedia (the first word means "uses"):
+As an example, this is the case on the following software template on the Dutch wikipedia (the first word means "uses", it can be found here: https://nl.wikipedia.org/wiki/Sjabloon:Infobox_software):
 
 ``` text
 {{Gebruikt Wikidata|P18|P154|P170|P178|P275|P277|P306|P348|P400|P548|P571|P577|P856|P1324|P2096}}

src/lib.rs

@@ -5,34 +5,72 @@ pub mod wikipedia_infobox_analyzer {
     use itertools::Itertools;
 
     /// Looks for a used template that does lists wikidata properties listing
-    pub fn extract_used_properties_from_template(template: String) -> Vec<String> {
+    pub fn extract_used_properties_from_template(template: String) -> Vec<u64> {
         let mut lines = template.lines();
         let used = lines
             .find(|line| line.starts_with("{{") && line.contains("Wikidata|"))
             .expect("Template should have a line declaring which properties are used");
 
         // The properties are listed inside the template with their number: P1|P2|P...
-        let re = Regex::new(r"P\d+").unwrap();
+        let re = Regex::new(r"P(\d+)").unwrap();
 
         // Find all matches and collect the properties into vector
-        re.find_iter(used)
+        re.captures_iter(used)
-          .map(|m| m.as_str().to_owned())
+          .map(|m| m[1].parse().unwrap())
           .collect()
     }
 
     /// Creates an ascii infobox out of the properties
-    pub fn format_infobox_from_used_properties(name: String, properties: Vec<String>) -> String {
+    pub fn format_infobox_from_used_properties(name: String, qid: String, properties_template: Vec<u64>, properties_item: Vec<u64>, ) -> String {
         let mut ascii_table = AsciiTable::default();
-        ascii_table.set_max_width(26);
+        ascii_table.set_max_width(80);
         ascii_table
             .column(0)
-            .set_header(name)
+            .set_header("Infobox requires")
             .set_align(Align::Left);
 
-        let data: Vec<Vec<&String>> = properties.iter().map(|v| vec![v]).collect();
+        ascii_table
+            .column(1)
+            .set_header(format!("Wikidata {name} ({qid})"))
+            .set_align(Align::Left);
+
+        // Display the property columns as a table
+        let data: Vec<Vec<String>> = join_columns(&properties_template, &properties_item)
+            .iter()
+            .map(|&m| {
+                  let (v1, v2) = m;
+                  vec![ v1.map_or("".to_owned(), |v| format!("P{v}").to_owned())
+                      , v2.map_or("".to_owned(), |v| format!("P{v}").to_owned())]
+                }
+            ).collect();
         ascii_table.format(data)
     }
 
+    /// Properties will be displayed in the right column if they are in the list for that column
+    /// All items in the left columns will be displayed
+    /// The remaining items in the right columns not in the left columns will trail in the right column
+    /// ┌───────┬───────┐
+    /// │   1   │   1   │
+    /// │   3   │       │
+    /// │       │   2   │
+    /// │       │   4   │
+    /// └───────┴───────┘
+    /// So for example, [1, 3] and [1, 2, 4] become [1, 2, 3, 4], as shown above
+    fn join_columns<T: Copy + Eq>(vec1: &[T], vec2: &[T]) -> Vec<(Option<T>, Option<T>)> {
+        vec1.iter()
+            .flat_map(|n1| {
+                if let Some(&n2) = vec2.iter().find(|&x| x == n1) {
+                    std::iter::once((Some(*n1), Some(n2)))
+                } else {
+                    std::iter::once((Some(*n1), None))
+                }
+            })
+            .chain(vec2.iter()
+                   .filter(|&n2| !vec1.contains(n2))
+                   .map(|n2| (None, Some(*n2))))
+            .collect()
+    }
+
     /// Fetches the wikidata label for the wiki property
     pub async fn fetch_name_for_wiki_property(pid: u64) -> String {
         let query = format!("
@@ -116,9 +154,9 @@ mod tests {
         assert_eq!(
             extract_used_properties_from_template(contents),
             vec![
-                "P18", "P154", "P1128", "P4103", "P2139", "P2295", "P2226", "P856", "P169",
+                18, 154, 1128, 4103, 2139, 2295, 2226, 856, 169,
-                "P1448", "P159", "P749", "P355", "P17", "P1056", "P452", "P576", "P112", "P127",
+                1448, 159, 749, 355, 17, 1056, 452, 576, 112, 127,
-                "P856", "P2096"
+                856, 2096
             ]
         );
 
@@ -128,8 +166,8 @@ mod tests {
         assert_eq!(
             extract_used_properties_from_template(contents),
             vec![
-                "P18", "P154", "P170", "P178", "P275", "P277", "P306", "P348", "P400", "P548",
+                18, 154, 170, 178, 275, 277, 306, 348, 400, 548,
-                "P571", "P577", "P856", "P1324", "P2096"
+                571, 577, 856, 1324, 2096
             ]
         );
     }
@@ -139,16 +177,22 @@ mod tests {
         assert_eq!(
             format_infobox_from_used_properties(
                 "Earth".to_string(),
-                vec!["P31".to_string(), "P361".to_string(), "P571".to_string()]
+                "Q2".to_string(),
+                vec![18, 170, 571],
+                vec![18, 31, 138, 170, 361, 571]
             ),
+
             concat![
-                "┌───────┐\n",
+                "┌──────────────────┬─────────────────────┐\n",
-                "│ Earth │\n", // item identifier: Q2
+                "│ Infobox requires │ Wikidata Earth (Q2) │\n",
-                "├───────┤\n",
+                "├──────────────────┼─────────────────────┤\n",
-                "│ P31   │\n", // property: instance of
+                "│ P18              │ P18                 │\n", // property: image
-                "│ P361  │\n", // property: part of
+                "│ P170             │ P170                │\n", // property: creator
-                "│ P571  │\n", // property: inception
+                "│ P571             │ P571                │\n", // property: inception
-                "└───────┘\n"
+                "│                  │ P31                 │\n", // property: instance of
+                "│                  │ P138                │\n", // property: inception
+                "│                  │ P361                │\n", // property: part of
+                "└──────────────────┴─────────────────────┘\n",
             ]
         );
     }

src/main.rs

@@ -15,14 +15,23 @@ fn main() -> Result<(), Box<dyn std::error::Error>> {
     let args = Cli::from_args();
 
     // Input to the program
-    let qid = fetch_wiki_item_by_article_title(
+    let title = args.wikipedia_article_title;
-        args.wikipedia_article_title,
+    let language = args.wikipedia_language_code;
-        args.wikipedia_language_code,
+    let template = std::fs::read_to_string(&args.infobox_template_file)?;
-    );
-    let _template_file = std::fs::read_to_string(&args.infobox_template_file)?;
 
-    println!("Qid: {qid}");
+    let qid = fetch_wiki_item_by_article_title(title.clone(), language.clone());
-    println!("Template file exists");
+    let properties_template = extract_used_properties_from_template(template);
+
+    let properties_item = fetch_properties_for_wiki_item(qid);
+    let table = format_infobox_from_used_properties(
+        title.clone(),
+        format!("Q{qid}"),
+        properties_template,
+        properties_item,
+    );
+
+    // Output of the program
+    println!("{table}");
 
     Ok(())
 }