On 06/14/2013 03:59 PM, Matt D wrote:
On 06/14/2013 03:14 PM, Dave Angel wrote:
On 06/14/2013 10:48 AM, Matt D wrote:
Hey,
here is a snip of my code.
#logger code----------------------------------------------
# first new line
#self.logfile.write('\n')
# date and time
#self.logfile.write('%s,'%(str(strftime("%Y-%m-%d %H:%M:%S",
gmtime()))))
# blah = [time, 'nac', 'tgid', 'source', 'dest', 'algid'] is
what we want
tmplist = []
tmplist.append(str(strftime("%Y-%m-%d %H:%M:%S", localtime())))
tmplist.append(field_values["nac"])
tmplist.append(field_values["tgid"])
tmplist.append(field_values["source"])
tmplist.append(field_values["dest"])
tmplist.append(field_values["algid"])
tmplist is an unnecessary complication. Did you look at my sample loop,
which I'll repeat here with a correction:
for k in FIELD_LIST_NAMES:
# get the value of the current TextCtrl field
f = field_values.get(k, None)
if not f is None:
#output the value with trailing comma
self.logfile.write('%s,'%(str(f)))
else:
self.logfile.write(",")
self.logfile.write("\n")
This code preserves your original feature of not crashing when the C++
program fails to fill in all your expected keys. It also makes sure
there will be unadorned commas for missing fields, making it possible
for a spreadsheet to read the columns correctly.
If you want to populate a list first, by all means do so, but do it in a
loop, using the FIELD_LIST_NAMES as keys.
One thing I didn't handle was the date field. I'd do that by adding it
to the field_values dict, or to a copy of it. That way, it's all
consistent, even though one field comes from local and the rest from the
pickle.
yes acutally this templist business broke my code. the TectCtrls in the
traffic panel would were not being populated and the logfile.csv was
empty.
So should i replace:
#logger code---------------
# first new line
self.logfile.write('\n')
# date and time
self.logfile.write('%s,'%(str(strftime("%Y-%m-%d %H:%M:%S",
localtime()))))
# loop through each of the TextCtrl objects
for k,v in self.fields.items():
# get the value of the current TextCtrl field
f = field_values.get(k, None)
if f:
# output the value with trailing comma
self.logfile.write('%s,'%(str(f)))
#end logger code ----------------
With the code you posted above?
Don't replace anything till you understand it. But if you think you do
then my code replaces the part starting at the for loop.
I am pretty sure that the reason i don't get the 'source' and 'dest'
fields is because of this:
#if the field 'duid' == 'hdu', then clear all the fields
if field_values['duid'] == 'hdu':
self.clear()
since the 'source' and 'dest' are in the LUD1 and not the HDU so it
doesn't update when the LDU1 comes through (if the LDU1) does actually
get serialized.
I don't know anything about LUDI or HDU. But perhaps you're saying that
some fields aren't in every pickle, but the value in the csv for each
line should be the last one pickled. In that case, you have a big logic
flaw in your code. When you output your stuff to the logfile, you use
only the values in field_values, not any values previously stored in
self.fields. Do you perhaps mean that whenever a value is missing from
the pickle, you want to use the one from self.fields?
If you happened to want exactly this, you could add the two middle lines
as below. something like:
f = field_values.get(k, None)
if f is None: #add me
f = self.fields.get(k, None) #add me
if not f is None:
But clearly, that's much more specific than you've ever been. There are
also better ways to do it if that's exactly what you want.
still haven't found a way to get to view the serialized
data.
print field_values, right at the beginning of update(). Or you could
pretty it up, by looping through its items().
--
DaveA
_______________________________________________
Tutor maillist - [email protected]
To unsubscribe or change subscription options:
http://mail.python.org/mailman/listinfo/tutor
