MapScript classObj Error

[geographika]

The following code

>>> import mapscript
>>> new_class = mapscript.classObj()

using the Python SWIG bindings version 1.3.39 produces this error:

Traceback (most recent call last):
  File "<pyshell#1>", line 1, in <module>
    new_class = mapscript.classObj()
  File "C:/Mapserver/bin/ms/python\mapscript.py", line 828, in __init__
    if args and len(args)!=0:
NameError: global name 'args' is not defined

Looking at the MapScript SWIG bindings the args parameter should not be in the constructor:

    def __init__(self, layer = None): 
        this = _mapscript.new_classObj(layer)
        try: self.this.append(this)
        except: self.this = this
        if 1:
        	   if args and len(args)!=0:
        		self.p_layer=args[0]
        	   else:
        		self.p_layer=None

[geographika]

Same issue for layerObj. See ​http://gis.stackexchange.com/questions/8735/using-mapscript-to-create-a-mapfile-from-scratch for more details.

[geographika]

Patch to custom Python SWIG file

[geographika]

This seems to be caused by custom Python code added to the pymodule.i file. In the attached patch I commented this out and the mapscript.py file works again.

[unicoletti]

I have committed some changes that should fix this issue in r12039 Please report if it's working.

[unicoletti]

no news, closing

[rblazek]

I have applied r12039 to 6.0.1 and with Python 2.5 it generates

def __init__(self, *args):
    this = _mapscript.new_layerObj(*args)
    try: self.this.append(this)
    except: self.this = this
    if 1:
        if map:
            self.p_map=map
        else:
            self.p_map=None

which obviously gives an error when executed. With Python 2.6 it is OK.

[unicoletti]

As much as I'd like to fix this I don't happen to have python 2.5 on any of my systems and I can't find a way to work around it (ie with conditional processing) and believe me I *did* try.

I'm sorry, but unless someone provides a patch for it I'll have to close it as WONTFIX.

[rblazek]

I have attached a patch which is working with python 2.5 and 2.6. I don't say that it is the best solution.

I believe that WONTFIX should only be used for 'features', i.e. something which is considered by developers to be a correct behaviour. Not for obvious bug.

[rblazek]

Support for both python 2.5 and 2.6 (version 6.0.1)

[unicoletti]

Patch from rblazek updated against trunk

[unicoletti]

Replying to rblazek:

I have attached a patch which is working with python 2.5 and 2.6. I don't say that it is the best solution.

Have tested your patch on 2.6 and generated a fresh one against trunk. Let me now if it works and I'll push it upstream asap.

I believe that WONTFIX should only be used for 'features', i.e. something which is > considered by developers to be a correct behaviour. Not for obvious bug.

[guillaume]

worked for me with python 2.6 Thanks

[unicoletti]

Committed in r13182

[aboudreault]

I've backported the fix in branch 6.0 in r13288.