[reportlab-users] new-para bug

Chad Miller chad_miller at symantec.com
Fri Jul 29 14:21:23 EDT 2005


I think I discovered a bug in the new Para.  A subtle, single character 
caused my message of 24 June (under a different e-mail address).

http://two.pairlist.net/pipermail/reportlab-users/2005-June/004065.html



--- reportlab/platypus/para.py 2004-01-21 03:33:05.000000000 -0500
+++ para.py     2005-07-29 13:58:14.149006389 -0400
@@ -1268,21 +1268,21 @@
             leading = self.style1.leading
         program = self.program
         self.cansplit = 1 # until proven otherwise
         if state:
             p.resetState(state)
             p.x = 0
             p.y = 0
             needatleast = state["leading"]
         else:
             needatleast = self.style1.leading
-        if availableHeight<=needatleast:
+        if availableHeight < needatleast:
             self.cansplit = 0
             #if debug:
             #    print "CANNOT COMPILE, NEED AT LEAST", needatleast, 'AVAILABLE', availableHeight
             return (availableHeight+1, availableWidth) # cannot split
         if parsedText is None and program is None:
             raise ValueError, "need parsedText for formatting"
         if not program:
             self.program = program = self.compileProgram(parsedText)
         if not self.formattedProgram:
             (formattedProgram, remainder, \


More information about the reportlab-users mailing list