Changeset 842

Timestamp:

06/15/07 15:41:33 (2 years ago)

Author:

thomase

Message:

All certainty measures, plus name-wise references.

Files:

• BADataMunger/trunk/placesaver.py (modified) (3 diffs)
• BADataMunger/trunk/tableparser.py (modified) (11 diffs)

BADataMunger/trunk/placesaver.py

@@ -83 +83 @@
-        tag_bibl = etree.Element("{%s}bibl" % TEI)
-        try:
-
+.replace("/ ", "/")
-        except:
-            tag_bibl.text = u"BAtlas %s %s" % (self.map_number, place.grid)
@@ -141 +141 @@
-            tag_css.append(tag_csn)
-            tag_cs.append(tag_css)
-
+
+
+
+
+
+
+
+
+
+
+
+
-            tag_nassoc = etree.Element("{%s}nameAssociation" % AWMC)
-            tag_nassoc.attrib['ref'] = pn.certainty
@@ -157 +168 @@
-                tag_tp.append(tag_tpn)
-                tag_fn.append(tag_tp)
+                
+            # secondary references for the name
+            if len(pn.references) > 0:
+                tag_refs = etree.Element("{%s}secondaryReferences" % AWMC)
+                for ref in pn.references:
+                    tag_bibl = etree.Element("{%s}bibl" % TEI)
+                    tag_bibl.text = ref
+                    tag_refs.append(tag_bibl)
+                    
+                tag_fn.append(tag_refs)
+
-                
-            ge.append(tag_fn)

BADataMunger/trunk/tableparser.py

@@ -211 +211 @@
-        elif d.tag == 'placenames':
-            
-
+ and references
-            try:
-                context = d.xpath("ancestor::when | ancestor::otherwise")[0]
@@ -221 +221 @@
-                except:
-                    pass
+                try:
+                    rcelli = context.xpath("./descendant::references")[0].attrib['cell']
+                except:
+                    pass
-            else:
-                try:
@@ -226 +230 @@
-                except:
-                    pass
-
-
+
+
+
+
+
+
+
-                self.placenames = self.parse_placenames(cells[celli], cells[pcelli])
+            elif rcelli:
+                self.placenames = self.parse_placenames(cells[celli], None, cells[rcelli])
-            else:
-                self.placenames = self.parse_placenames(cells[celli])
@@ -263 +274 @@
-        self.periods = self.parse_periods(cells[2])
-        self.locdesc = normalizetext(getalltext(cells[3]))
-
+, cells[4]
-        self.namestring = normalizetext(getalltext(cells[1]))
-        self.references = self.parse_references(cells[4])
@@ -271 +282 @@
-        # name | references | comment
-        self.type = 'false'
-
+, None, cells[1]
-        self.namestring = normalizetext(getalltext(cells[0]))
-        self.references = self.parse_references(cells[1])
@@ -301 +312 @@
-        self.periods = self.parse_periods(cells[1])
-        self.locdesc = normalizetext(getalltext(cells[2]))
-
+, cells[3]
-        self.namestring = normalizetext(getalltext(cells[0]))
-        self.references = self.parse_references(cells[3])
@@ -343 +354 @@
-        return [normalizetext(place) for place in places]
-        
-
+, refcell=None
-        """We can have multiple placenames in a single cell, and there are two types of delimiters.
-        Parse this mess to get the individual names, and then deal with variants too."""
@@ -388 +399 @@
-                pname.periods = self.periods
-                
+        # associate references, if necessary, with the placenames
+        # if separate references are provided for each work 
+        if refcell:
+            refparas = refcell.xpath("*[local-name()='p']")
+            if len(refparas) > 1:
+                for pname in placenames:
+                    try:
+                        pname.references = self.parse_references(refparas[pname.originalPosition])
+                    except:
+                        pass
+            elif len(self.references) > 0:
+                for pname in placenames:
+                    pname.references = self.references
+
-        return placenames
-        
@@ -429 +454 @@
-        m = re.match(pattern, ntext)
-        if m:
-t
+f
-            pattern = u'[\[\]]'
-            newntext = re.sub(pattern, u'', ntext)
@@ -447 +472 @@
-        references=[]
-        
-
+descendant-or-self::
-        for paragraph in paragraphs:
-            ptext = normalizetext(getalltext(paragraph))
@@ -475 +500 @@
-        
-        
-    def parse_periodcell(self, periodcell):
-        print "parse_periodcell"
-        periods=[]
-        pctext = normalizetext(getalltext(periodcell))
-        if len(pctext) > 0:
-            periodishes = periodcell.xpath("descendant::*[local-name()='p']")
-            try:
-                periodish = periodishes[self.originalPosition]
-            except:
-                pass
-            if periodish:
-                lpstring = normalizetext(getalltext(periodish))
-                if len(lpstring) > 0:
-                    for pcode in periodcodes:
-                        qpcode = '%s?' % pcode
-                        if lpstring.find(qpcode) != -1:
-                            lpstring = lpstring.replace(qpcode, '')
-                            if lpstring.find(pcode) != -1:
-                                lpstring = lpstring.replace(pcode, '')
-                                periods.append((pcode, 'confident'))
-                            else:
-                                periods.append((pcode, 'less-confident'))
-                        else:
-                            if lpstring.find(pcode) != -1:
-                                lpstring = lpstring.replace(pcode, '')
-                                periods.append((pcode, 'confident'))
-                    
-        return periods
-        
-    def __str__(self):