Re: [Rest2web-develop] RE: File Bug

[Fuzzyman <fuz...@vo...>]

Andrew Ittner wrote:
> Here is some debug output for the root's index:
> self.dir=; filename=index.txt; entry=rest2web140x62.gif;
> src=index.txt/rest2web140x62.gif
>
>   
So this works fine.

> And for projects/rest2web's index:
> self.dir=projects/rest2web; filename=projects/rest2web/index.txt;
> entry=halloffamelogo.png;
> src=projects/rest2web/projects/rest2web/index.txt/halloffamelogo.png
>
> This line:
> src = join(self.dir, dirname(filename), entry)
>
> will concatenate the current directory, AND the current directory of a file
> in the current directory, and THEN the filename.  So you end up with
> DIR/SUBDIR/DIR/SUBDIR/filename.ext.
>
>   
Yup - I understand.

I think the method ``get_real_restindex`` is doing this to the filename 
- so I will work out a generic method of solving this. I wonder why this 
worked for me though ?

> Since you want to merge these 2 loops, perhaps I should wait to add a fix.
> Here's what I recommend: compare self.dir and dirname(filename)'s results:
> if they match, use only one.  If they don't, use one.  I think we need a
> complete test suite on this one, because you talked about setting the
> index-file option, which I do not use in combination with the file keyword.
>   
I am not currently using the file keyword at all, I do however use the 
index-file keyword quite a lot, so it is important to me that this works 
correctly.

I will sort it out.

Thanks for your help in diagnosing this.

All the best,

Fuzzyman
http://www.voidspace.org.uk/python/index.shtml

> Andrew
>
> -----Original Message-----
> From: res...@li...
> [mailto:res...@li...]On Behalf Of Michael
> Foord
> Sent: Thursday, April 06, 2006 12:15 AM
> To: res...@li...
> Subject: Re: [Rest2web-develop] RE: File Bug
>
>
>
>
>
> On 06/04/06, Andrew Ittner <and...@us...> wrote:
> I love one-line fixes.
>
> Changed
>     src = join(self.dir, filename, entry)
>
>
> But in SVN I had already changed that line to :
>
>     src = join(self.dir, dirname(filename), entry)
>
> ??? Because I'd like to move the two chunks of code into a single function,
> that dirname will be needed for the case of index files.
>
> I'll make that change and you can test again. :-)
>
> Fuzzyman
> http://ww.voidspace.org.uk/python/index.shtml
>
>
>
> to
>     src = join(self.dir, entry)
>
> It appears that the filename value set the path to
> "index.txt /thefilename.ext" instead of just "thefilename.ext".  Once I
> removed it, both the index and non-index pages copied files fine (so I did
> not need to change the other loop's src line).
>
> Andrew
>
>   
>> -----Original Message-----
>> From: Fuzzyman [mailto:fuz...@vo...]
>> Sent: Tuesday, April 04, 2006 2:34 AM
>> To: res...@li...; Andrew Ittner
>> Subject: Re: File Bug
>>
>>
>> Hello Andrew,
>>
>> I would certainly appreciate your eyes in tracking down this bug. I
>> don't *think* it can be platform specific though, as there is no
>> platform specific code (I use platform independent functions from
>> ``os.path`` and ``shutil`` to do all the work).
>>
>> I can see where the bug is happening - for you it seems to be where the
>> location of the source file is determined. I've pasted the full code in
>> below - and explained the logic of the first part which determines the
>> source location. You could add in a print statement to see where it
>> thinks the source file *should be*, which will give us a clue as to what
>> is happening.
>>
>> The code exists twice (which means it ought to be turned into a method
>> of Processor). Once in the ``process`` method and once in the
>> ``processindex`` method.
>>
>> When rest2web starts, it changes the current directory to be the 'start'
>> directory (top level of the site directory tree).
>>
>> As it goes through the files, the current directory being worked on is
>> stored in ``self.dir`` (self being the processor instance). At the top
>> level directory the following is set in ``__init__``, ``self.dir =
>> os.curdir``. (Is it possible that this is the cause of the problem ?)
>>
>> Following code taken from -
>> http://rest2web.python-hosting.com/file/trunk/rest2web/restprocessor.py
>> - from the process method.
>>
>> # also copy files (from restindex) into same dir as target
>> for entry in restindex['file']:
>>     src = join(self.dir, dirname(filename), entry)
>>     ## XXX Add a ``print src`` here :-)
>>     if not os.path.isfile(src):
>>         print ('File "%s" referenced by file keyword missing.' %
>>             entry)
>>         continue
>>     #
>>     target_file_name = url2pathname(target)
>>     targetfile = os.path.normpath(join(
>>         self.target, self.dir, target_file_name))
>>     dest = join(dirname(targetfile), split(entry)[1])
>>     comparefile = os.path.normpath(join(
>>         self.compare, self.dir, target_file_name))
>>     cmp = join(dirname(comparefile), split(entry)[1])
>>     if comparefiles(src, cmp):
>>         print 'File "%s" identical, skipping.' % entry
>>         continue
>>     if not isdir(dirname(dest)):
>>         os.makedirs(dirname(dest))
>>     print 'Copying "%s".' % entry
>>     copy2(src, dest)   # alias for shutil.copy2
>>
>> I'll step through the code to explain the process.
>>
>>     for entry in restindex['file']:
>>
>> restindex['file'] should be a list of all your file keywords from the
>> page being processed.
>>
>>
>>     src = join(self.dir, dirname(filename), entry)
>>
>> ``join`` is an alias for ``os.path.join``.
>>
>> ``filename`` is the name of the file being processed. When we are
>> processing an index file, it could actually be a path relative to
>> ``self.dir`` if the ``index-file`` option is set.
>>
>> You were getting lots of  ``File "..." referenced by file keyword
>> missing.`` errors - so this logic seems to be faulty ?
>>
>> If you add the print statement I suggest, then you should be able to
>> tell what is happening. I'm sure the solution is simple enough - but I
>> can't see it.
>>
>> Thanks
>>
>> Fuzzyman
>> http://www.voidspace.org.uk/python/index.shtml
>>     
>
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by xPML, a groundbreaking scripting language
> that extends applications into web and mobile media. Attend the live webcast
> and join the prime developer group breaking into this new coding territory!
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
> _______________________________________________
> Rest2web-develop mailing list
> Res...@li...
> https://lists.sourceforge.net/lists/listinfo/rest2web-develop
>
>