Re: some try-ark suggestions

On 10/6/01 19:31, [email protected] wrote:

>After copying over my myteam1 directory (based on the sample1 directory)
>from yesterday's attempt, I get this:
>
>[berto@localhost myteam1]$ try-ark/try-ark | tee /tmp/try-ark.out
>try-ark/try-ark: warning: this script may be overly paranoid and, um, 
>wrong...
>+ ./arkcmd package reveal --verbose --use-deps=max arkbase
>Traceback (most recent call last):
>  File "./arkcmd", line 79, in ?
>    main()
>  File "./arkcmd", line 66, in main
>    things_mgr.doToolSubcommand()
>  File "./ark/thing.py", line 175, in doToolSubcommand
>    victims       = self.unpackSpecs(ark_ctrl.cmdLineNonOptions())
>  File "./ark/thing.py", line 108, in unpackSpecs
>    thing = self.lookup(thing_name,team_id)
>  File "./ark/thing.py", line 69, in lookup
>    team_id = ark.team.ACTING_TEAM().id
>  File "./ark/team.py", line 55, in ACTING_TEAM
>    _ActingTeam.instance = ArkTeam(acting_team_id)
>  File "./ark/team.py", line 135, in __init__
>    (self._is_prototype, xml_version, self._xmlf) = 
>xml_file.read(explicit_filename="%s/team.xml" % 
>(ark.control.ArkControl().repository(team_id)))
>  File "./ark/xmlfile.py", line 152, in read
>    root_elem.normalize()
>  File "/usr/lib/python2.0/xml/dom/minidom.py", line 74, in __getattr__
>    raise AttributeError, key
>AttributeError: normalize
>
>Am I getting any warmer?

Damn. I saw a comment in the code about this, but I couldn't reproduce 
the problem. Must be a difference between Python 2.0 and 2.1. Looks like 
the Python/XML world has it in for me ;-)

>On the other hand, if I and/or the PyXML package maintainers "botched" 
>things,
>it could happen to anyone.  What's the correct response to this?  Just tell
>me (and many others?) to go fix my (our) botched installation(s)?  Or for
>the Arusha code to be robust enough to deal with these messes?
>
>I favor your continuing to hack a software solution to this problem.  ;-)

I'm going to continue to bash on a working Python 2.x solution so we can 
skip the PyXML thing. In the meantime I'll figure out who to speak to in 
the PyXML team about the packaging.


Then, just before I could send this email, Berto wrote:

>It would be a reasonable response, telling me (and others) just to go fix
>the botched installation (for if it's broken for Arusha, it's broken for
>other packages as well).  If so, you should add some cautions to the
>Arusha install notes, e.g., "don't use the current SourceForge PyXML RPM,
>because it's broken" or "do an rpm -qlp of the RPM to check..." or, "before
>attempting try-ark, check the following things in an interactive python
>session first" or ...

Probably reasonable, but I've got the bit between my teeth now ;-)

>Do I read you correctly that python 2.0 has the needed xml pieces there by
>default, so that no separate PyXML installation may be necessary?

Yeah it's supposed to, but the minidom DOM implementation seems to be a 
bit scaled down from the full PyXML stuff. I'm not sure the whole 
politics there and why PyXML hasn't made it into the full distribution - 
it's been stable for a while.

>(Sigh, I sense I'll have to dive into finally learning Python after all.)

I wouldn't want to stop you doing that :-)

:-j


-- 
jonathan hogg, one good idea ltd, 131 queen margaret dr., glasgow g20 8pd
http://www.onegoodidea.com/ tel:+44-(0)7976-614338 fax:+44-(0)7970-537451