You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@commons.apache.org by Rory Winston <rw...@eircom.net> on 2006/08/30 10:21:57 UTC

Re: [Fwd: Re: apache bugzilla 39109 ... assignee very quite ...]

Henrik

I think I see the problem with the current setup.

1. Firstly, the parser needs to "initialize" itself via an FTP listing
header line in preParse(), in order to set the isType flag and the
parent class regex. As long as we have a way of explicitly setting this,
we can get the class to work in a unit test.
2.Secondly, the current parser will never be able to parse a line such as:

Migrated                                                file1.I

Because as far as the parser is concerned, this is not a valid line, it 
is incomplete. The old parser used a much simpler regex that was able to 
handle lines like this, but the old parser simply extracted the last 
token (the filename) and used that. Your parser is able to extract more 
information - when it exists. If it is possible for a line such as the 
above to occur in practise (which I am assuming it is) then we need to 
be able to handle it as well.

Also, the method parseFileList() expects to find a valid value for the 
"DSORG" header field. If it does not, it returns fakse from 
parseFileList(), faiing the parse. This will not work for lines where a 
"DSORG" value may not be present, so we would need a way of handling 
this as well.

I think we are nearly there -if we can get these issues sorted, then we 
can look at getting this in to the next release.

PS I am forwarding this message back on to the dev list, as we may be 
able to get some useful input there too.

Thanks
Rory





---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org