Merge branch 'NSoiffer:main' into main

Author: MartheGjelstad

PythonScripts/translate-unicode.py

@@ -404,39 +404,64 @@ def print_set(name, set, orig_dict):
 # Strangely, if you copy the '.../en/definitions.yaml' and paste that into translate.google.com, it does it all.
 # Rather than waste a bunch more time on this, the code assumes you've translated the file already and stored
 #   it in 'google-defs.yaml' in the current dir
-# It then goes through the English version leaving the English and  pulling out only the translated *values*
+# It then goes through the English version leaving the English and pulling out only the translated *values*
 #   from 'google-defs.yaml' writing '[lang]-definitions.yaml'.
 def translate_definitions(path_to_mathcat: str, lang: str):
     if lang == 'nb' or lang == 'nn':
         lang = 'no'  # google doesn't know those variants
 
-    file_to_translate = "{}/Rules/Languages/en/definitions.yaml".format(path_to_mathcat)
+    file_to_translate = f'{path_to_mathcat}/Rules/Languages/en/definitions.yaml'
+    translated_file = f'{path_to_mathcat}/Rules/Languages/{lang}/definitions.yaml'
     with open("google-defs.yaml", 'r', encoding='utf8') as google_stream:
         translated_lines = google_stream.readlines()
         with open(file_to_translate, 'r', encoding='utf8') as in_stream:
-            with open(f"{lang}/definitions.yaml", 'w', encoding='utf8') as out_stream:
+            with open(translated_file, 'w', encoding='utf8') as out_stream:
                 lines = in_stream.readlines()
-                i = 0
+                i_en = 0
+                i_trans = 0
                 n_lines = len(lines)
-                while i < n_lines:
-                    if not (lines[i].startswith('#')) and lines[i].find(': [') >= 0:
+                while i_en < n_lines:
+                    if not (lines[i_en].startswith('#')) and (lines[i_en].find(': [') >= 0 or lines[i_en].find(': {') >= 0):
                         # handles 'xxx: [' inclusive of the line with the matching ']'
-                        i = translate_definition(i, lines, translated_lines, out_stream)
+                        (i_en, i_trans) = translate_definition(i_en, lines, i_trans, translated_lines, out_stream)
                     else:
-                        out_stream.write(lines[i])
-                    i += 1
-
-
-def translate_definition(start: int, lines: list[str], translated_lines: list[str], out_stream):
-    out_stream.write(lines[start])
-    i = start+1    # first line is 'name: ['
-    while i < len(lines):
-        if lines[i].find(']') >= 0:
-            out_stream.write(lines[i])
-            return i
-        out_stream.write(translated_lines[i].replace("“", "'").replace("”", "'").replace("、", ","))  # Chinese
-        i += 1
-    return i
+                        out_stream.write(lines[i_en])
+                    i_en += 1
+                    # i_trans += 1
+
+
+def translate_definition(i_en: int, lines: list[str], i_trans: int, translated_lines: list[str], out_stream) -> (int, int):
+    out_stream.write(lines[i_en])
+    i_en = i_en+1    # first line is 'name: [' or 'name: {'
+    # sync lines -- find '[' in translation
+    while not (translated_lines[i_trans].find(': [') >= 0 or translated_lines[i_trans].find(': {') >= 0):
+        i_trans += 1
+    i_trans += 1   # skip the [/{ line
+    while i_en < len(lines):
+        if (
+            translated_lines[i_en].strip().startswith('#') or
+            len(translated_lines[i_en].strip()) == 0 and len(translated_lines[i_trans].strip()) == 0
+        ):
+            out_stream.write(lines[i_en])
+        elif lines[i_en].find(']') >= 0 or lines[i_en].find('}') >= 0:
+            out_stream.write(lines[i_en])
+            return (i_en, i_trans)
+        elif len(translated_lines[i_trans].strip()) == 0:  # google sometimes adds blank lines
+            i_trans += 1
+            continue
+        else:
+            print(f'en: {lines[i_en].strip()}\ntr: {translated_lines[i_trans].strip()}')
+            # get indentation right
+            i_spaces = lines[i_en].find('"')
+            cleaned_line = (
+                translated_lines[i_trans]
+                .replace("“", '"').replace("”", '"').replace("„", '"').replace("、", ",")  # Chinese
+                .lstrip()
+            )
+            out_stream.write(f'{" ".ljust(i_spaces)}' + cleaned_line)
+        i_en += 1
+        i_trans += 1
+    return (i_en, i_trans)
 
 
 def build_euro(lang: str):
@@ -502,7 +527,7 @@ def write_euro_braille_file():
 # (sre_only, mp_only, differ, same) = dict_compare("fr", get_sre_unicode_dict(SRE_Location, "fr"), get_mathplayer_unicode_dict(MP_Location, "fr"))
 # (sre_only, mp_only, differ, same) = dict_compare("it", get_sre_unicode_dict(SRE_Location, "it"), get_mathplayer_unicode_dict(MP_Location, "it"))
 
-language = "ru"
+language = "pl"
 # build_new_translation("..", language, "unicode")
 # build_new_translation("..", language, "unicode-full")
 

Rules/Languages/en/navigate.yaml

@@ -232,7 +232,7 @@
 - name: zoom-in-mrow-in-math
   # zooming in only once is meaningless because 'math' has only a single child and it was spoken at the math level -- dig inside and do it again 
   tag: math
-  match: "($NavCommand = 'ZoomIn' or $NavCommand = 'MoveNextZoom' or $NavCommand = 'MovePreviousZoom')"
+  match: "$NavCommand = 'ZoomIn' or $NavCommand = 'MoveNextZoom' or $NavCommand = 'MovePreviousZoom'"
   replace:
   - test:
       if: "$NavCommand = 'MovePreviousZoom'"
@@ -278,7 +278,7 @@
           variables: [Move2D: "'in'", Child2D: "*[1]/*[1]"]   # phrase('in' the denominator)
           replace: [x: "."]
       - set_variables: [NavNode: "*[1]/*[1]/@id"] # skip mtd
-    - else_if: "*[1][self::m:mrow and IsBracketed(., '(', ')', false) or IsBracketed(., '[', ']', false)]" # auto zoom
+    - else_if: "*[1][self::m:mrow and (IsBracketed(., '(', ')', false) or IsBracketed(., '[', ']', false))]" # auto zoom
       then:
       - with:
           variables: [Move2D: "'in'", Child2D: "*[1]"]   # phrase('in' the denominator)
@@ -459,7 +459,7 @@
       variables: [SayCommand: "string($NavVerbosity = 'Verbose')"]
       replace: [x: "."]
   - test:
-      if: "$NavMode='Enhanced' and parent::*[self::m:mrow and IsBracketed(., '(', ')', false) or IsBracketed(., '[', ']', false)]"
+      if: "$NavMode='Enhanced' and parent::*[self::m:mrow and (IsBracketed(., '(', ')', false) or IsBracketed(., '[', ']', false))]"
       then: [x: ".."] # auto-zoom: move out a level and retry
       else:
       - with:
@@ -1659,7 +1659,7 @@
   - t: "inside"                                           # phrase('inside' a big expression)
   - pause: medium
   - test:
-    - if: "$NavMode='Enhanced' and parent::*[self::m:mrow and IsBracketed(., '(', ')', false) or IsBracketed(., '[', ']', false)]"
+    - if: "$NavMode='Enhanced' and parent::*[self::m:mrow and (IsBracketed(., '(', ')', false) or IsBracketed(., '[', ']', false))]"
       then: [x: ".."] # auto-zoom up
     - else_if: "$NavCommand = 'WhereAmI'"
       then: [set_variables: [NavNode: "@id"]]

src/chemistry.rs

@@ -789,11 +789,12 @@ fn likely_chem_equation(mathml: Element) -> isize {
     // debug!("start likely_chem_equation:\n{}", mml_to_string(mathml));
 	// mrow -- check the children to see if we are likely to be a chemical equation
 
-    // concentrations should either be unscripted or have a superscript
+    // concentrations should either be unscripted or have a superscript that isn't a charge
     // they occur in mrows or mfracs
     if IsBracketed::is_bracketed(mathml, "[", "]", false, true) {
         let parent_name = name(get_parent(mathml));
-        if parent_name == "mfrac" || parent_name == "mrow" || parent_name == "msup" || parent_name == "math" {
+        if parent_name == "mfrac" || parent_name == "mrow"  || parent_name == "math" || 
+           (parent_name == "msup" && likely_chem_superscript(as_element(mathml.following_siblings()[0])) < 0){
             return if as_element(mathml.children()[0]).attribute(CHEM_FORMULA).is_some() {CHEMISTRY_THRESHOLD}  else {NOT_CHEMISTRY};
         }
     }
@@ -2294,6 +2295,7 @@ mod chem_tests {
 
     #[test]
     fn dichlorine_hexoxide() {
+        init_logger();
         let test = "<math><mrow>
             <msup>
             <mrow><mo>[</mo><mi>Cl</mi><msub><mi>O</mi><mn>2</mn></msub><mo>]</mo></mrow>

src/speech.rs

@@ -661,23 +661,21 @@ impl Intent {
         /// "lift" up the children any "TEMP_NAME" child -- could short circuit when only one child
         fn lift_children(result: Element) -> Element {
             // debug!("lift_children:\n{}", mml_to_string(result));
-            result.replace_children(
-                result.children().iter()
-                    .map(|&child_of_element| {
-                        match child_of_element {
-                            ChildOfElement::Element(child) => {
-                                if name(child) == "TEMP_NAME" {
-                                    assert_eq!(child.children().len(), 1);
-                                    child.children()[0]
-                                } else {
-                                    child_of_element
-                                }
-                            },
-                            _ => child_of_element,      // text()
+            // most likely there will be the same number of new children as result has, but there could be more
+            let mut new_children = Vec::with_capacity(2*result.children().len());
+            for child_of_element in result.children() {
+                match child_of_element {
+                    ChildOfElement::Element(child) => {
+                        if name(child) == "TEMP_NAME" {
+                            new_children.append(&mut child.children());  // almost always just one
+                        } else {
+                            new_children.push(child_of_element);
                         }
-                    })
-                    .collect::<Vec<ChildOfElement>>()
-            );
+                    },
+                    _ => new_children.push(child_of_element),      // text()
+                }
+            }
+            result.replace_children(new_children);
             return result;
         }
     }    
@@ -983,7 +981,7 @@ pub struct MyXPath {
 
 impl fmt::Display for MyXPath {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
-        return write!(f, "x: \"{}\"", self.rc.string);
+        return write!(f, "\"{}\"", self.rc.string);
     }
 }
 
@@ -1195,6 +1193,7 @@ impl MyXPath {
         return match result {
             Ok(val) => Ok( val ),
             Err(e) => {
+                debug!("MyXPath::trying to evaluate:\n  '{}'\n caused the error\n'{}'", self, e.to_string().replace("OwnedPrefixedName { prefix: None, local_part:", "").replace(" }", ""));
                 bail!( "{}\n\n",
                      // remove confusing parts of error message from xpath
                     e.to_string().replace("OwnedPrefixedName { prefix: None, local_part:", "").replace(" }", "") );
@@ -2706,7 +2705,7 @@ mod tests {
         assert_eq!(speech_pattern.tag_name, "math", "\ntag name failure");
         assert_eq!(speech_pattern.pattern.rc.string, ".", "\npattern failure");
         assert_eq!(speech_pattern.replacements.replacements.len(), 1, "\nreplacement failure");
-        assert_eq!(speech_pattern.replacements.replacements[0].to_string(), r#"x: "./*""#, "\nreplacement failure");
+        assert_eq!(speech_pattern.replacements.replacements[0].to_string(), r#""./*""#, "\nreplacement failure");
     }
 
     #[test]

src/xpath_functions.rs

@@ -753,7 +753,7 @@ struct BaseNode;
     /// Recursively find the base node
     /// The base node of a non scripted element is the element itself
     fn base_node(node: Element) -> Element {
-        let name = name(node);
+        let name = node.attribute_value(MATHML_FROM_NAME_ATTR).unwrap_or(name(node));
         if ["msub", "msup", "msubsup", "munder", "mover", "munderover", "mmultiscripts"].contains(&name) {
             return BaseNode::base_node(as_element(node.children()[0]));
         } else {