> diff --git a/iutil.py b/iutil.py
> index 39d398a..da97256 100644
> --- a/iutil.py
> +++ b/iutil.py
> @@ -1014,6 +1019,63 @@ def reIPL(anaconda, loader_pid):
> # or a two-item list with errorMessage and rebootInstr (=> shutdown)
> return message
>
> +def stringFileContents (filename):
> + try:
> + data = None
> +
> + fp = open (filename)
> +
> + data = fp.read ().strip ()
> +
> + fp.close ()
> +
> + dataContents = ""
> +
> + for ch in data:
> + if ch in string.printable:
> + dataContents += ch
> + else:
> + dataContents = None
> + break
> +
> + if dataContents is not None:
> + return dataContents
> +
> + dataContents = ""
> +
> + for ch in data:
> + dataContents += hex (ord (ch)) + " "
> +
> + return dataContents
> +
> + except Excception, e:
> +
> + return "ERROR"
Note the "Excception" typo above. Speaking of exceptions, I don't
really care for the style of just catching the top-level exception, as
this tends to have the unintended consequence of hiding unrelated
errors. Since this is a library function, I'd prefer to not handle
exceptions in it and let the caller deal with that.
Can you show me what the sample output from this function would be? I
find the two loops over data a little odd.
> +def dumpReIPLFiles ():
> + try:
> + filename = "/sys/firmware/reipl/reipl_type"
> +
> + if not os.path.isfile (filename):
> + return
> +
> + type = stringFileContents (filename)
> +
> + reIPL_log.info ("%s = %s" % (filename, type,))
> +
> + directory = "/sys/firmware/reipl/%s" % (type,)
> +
> + if not os.path.exists (directory):
> + return
> +
> + for directoryFile in os.listdir (directory):
> + contents = stringFileContents ("%s/%s" % (directory, directoryFile,))
> +
> + reIPL_log.info ("%s/%s = %s" % (directory, directoryFile, contents,))
> +
> + except Exception, e:
> + log.info("dumpReIPLFiles: Caught exception %s", (e,))
The same comment about catching Exception applies here. Also I think
that you're really only going to have an exception coming up from
stringFileContents, since you check os.path.{isfile,exists} already.
Otherwise, this method looks fine.
- Chris
_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/anaconda-devel-list
