Add normalization mn tags in mml

convert2symLG/mml2lg.pl

@@ -9,7 +9,7 @@ no warnings;
 
 if($#ARGV < 0 or $#ARGV > 2){
 print "CROHMELib CROHME .inkml to Label Graph (.lg) Converter
-Copyright (c) H. Mouchère and R. Zanibbi, 2012-2014
+Copyright (c) H. Mouchère and R. Zanibbi, 2012-2023
 
 Usage:  mml2lg.pl [-s] fileIn.inkml [fileOut.lg]
 
@@ -18,7 +18,7 @@ standard output (STDOUT).
 The option -s produces the short format with 'O' and 'OE' lines
 instead of 'N' and 'E' lines.
 
-Version : 1.4
+Version : 1.5
 Authors: Harold Mouchère (LUNAM/University of Nantes/IRCCyN/IVC)
 	R. Zanibbi (DPRL Lab, RIT, USA)
 ";
@@ -46,7 +46,7 @@ my %tagToSRT = (
 	"msup" => [[0,1,'Sup']],
 	"msub" => [[0,1,'Sub']],
 	"mfrac" => [[-1,0,'Above'],[-1,1,'Below']],
-	"msqrt" => [[-1,0,'Inside'],[-1,1,'Inside'],[0,1,'Right']],
+	"msqrt" => [[-1,0,'Inside'],[0,1,'Right']],
 	"mroot" => [[-1,0,'Inside'],[-1,1,'Above']],
 	"munder" => [[0,1,'Below']],
 	"munderover" =>  [[0,1,'Below'], [0,2,'Above']],

convert2symLG/process_mml.py

@@ -53,6 +53,29 @@ def add_ID(mml):
             subtree['xml:id']='_'+str(count)
     return mml
 
+# normalize multi-character mn tags inserting mrow tags
+def norm_mn(mml):
+    soup = BeautifulSoup(open(mmlPath),'lxml')
+    for subtree in mml.findAll('mn'):
+        if len(subtree.text) > 1 :
+            #print ("mn tag with more than one character is found:", subtree)
+            d1 = subtree.text[0]
+            d2 = subtree.text[1:]
+            # rename as mrow tag
+            subtree.string = ""
+            subtree.name = 'mrow'
+            # add new tags mn for first digit and one for next digits
+            mnd1 = soup.new_tag('mn')
+            mnd1.string = d1
+            subtree.insert(0,mnd1)
+            next = soup.new_tag('mn')
+            next.string = d2
+            subtree.insert(1,next)
+            # recursive call to normalize the new mn tag
+            norm_mn(subtree)
+            #print("     => new tree :", subtree)
+    return mml
+
 #return mathml unicodes into latex labels
 def normalizeSymbol(label):
     if label in ['rightarrow', 'sum', 'int', 'pi', 'leq', 'sin','cos','tan', 'lim',
@@ -181,6 +204,7 @@ if __name__=='__main__':
         if soup('math'):
             body=soup('math')[0]
             clean_mml=remove_unknown_tags(body)
+            clean_mml = norm_mn(clean_mml)
             id_mml=add_ID(clean_mml)
         
             write_mml(id_mml,name,outDir)

tests/SymbLG/sqrt.txt

+$\sqrt{24+1234}$
\ No newline at end of file

tests/SymbLG/test.txt

+$\frac{15}{xy} \sin \theta ^{23}$
\ No newline at end of file