src/GrampsDbUtils/_WriteGedcom.py :#2008 fix bug in breakout()

src/GrampsDbUtils/test/gedwrite_breakup_test.py : unittest for breakout



svn: r10470

src/GrampsDbUtils/_WriteGedcom.py

@@ -252,11 +252,17 @@ def breakup(txt, limit):
     maximum length specified, while breaking words in the middle of a word
     to avoid issues with spaces.
     """
+    if limit < 1:
+        raise ValueError("breakup: unexpected limit: %r" % limit)
     data = []
-    while limit < len(txt)+1:
-        idx = limit-1
-        while txt[idx-1].isspace() or txt[idx].isspace() :
+    while len(txt) > limit:
+        # look for non-space pair to break between
+        idx = limit
+        while idx>0 and (txt[idx-1].isspace() or txt[idx].isspace()):
             idx -= 1
+        if idx == 0:
+            #no words to break on, just break at limit anyway
+            idx = limit
         data.append(txt[:idx])
         txt = txt[idx:]
     if len(txt) > 0:

src/GrampsDbUtils/test/gedwrite_breakup_test.py

@@ -0,0 +1,57 @@
+"""gedwrite_breakup_test tests the breakup function in GedcomWrite"""
+import unittest as U
+
+from test.test_util import msg, path_append_parent
+path_append_parent();
+import _WriteGedcom as WG
+
+class Test1(U.TestCase):
+    def test1a_common(s):
+        #txt, limit, [split-results]
+        L=4
+        dat = (
+            ##0123456 
+            ("a",  L, ["a",]),
+            ("abc",  L, ["abc",]),
+            ("abcd",  L, ["abcd",]),
+            ("abcde",  L, ["abcd","e"]),
+            ("1234567",  L, ["1234","567"]),
+            ("12345678",  L, ["1234","5678"]),
+            ("123456789",  L, ["1234","5678", "9"]),
+        )
+        for (t,l,r) in dat:
+            g = WG.breakup(t,l)
+            s.assertEquals(g,r, msg(g,r, "breakup(%r,%d) results" % (t,l)))
+
+    def test1b_spaces(s):
+        #txt, limit, [split-results]
+        L=4
+        dat = (
+            ##0123456 
+            ("a b ",  L, ["a b ",]),
+            (" a b",  L, [" a b",]),
+            ("asdf g", L, ["asd", "f g"]),
+            ("    z", L, ["    ", "z"]),
+            ("     z", L, ["    ", " z"]),
+            ("      A", 2, ["  ", "  ", "  ", "A"]),
+        )
+        for (t,l,r) in dat:
+            g = WG.breakup(t,l)
+            s.assertEquals(g,r, msg(g,r, "breakup(%r,%d) results" % (t,l)))
+
+    def test1c_unusual(s):
+        #just documenting behavior for unlikely input
+        #txt, limit, [split-results]
+        dat = (
+            ##0123456 
+            ("",  4, []),
+            ("xyz", 1, ["x", "y", "z"]),
+        )
+        for (t,l,r) in dat:
+            g = WG.breakup(t,l)
+            s.assertEquals(g,r, msg(g,r, "breakup(%r,%d) results" % (t,l)))
+        s.assertRaises(ValueError, WG.breakup, "xy",0)
+
+if __name__ == "__main__":
+    U.main()
+#===eof===