1. Rob Managan
  2. SCons_SONAME


SCons_SONAME / bin / xmlagenda.py

#!/usr/bin/env python

# Query the scons.tigris.org database for the issues of interest.
# The typical triage query is found on http://www.scons.org/wiki/BugParty

# Download the issues from Issuezilla as XML; this creates a file
# named 'issues.xml'.  Run this script in the dir containing
# issues.xml (or pass full path as arg to this script) to translate
# 'issues.xml' into a CSV file named 'editlist.csv'.  Upload the CSV
# into a Google spreadsheet.

# In the spreadsheet:
# Select all the columns and pick "align-->top"
# Select the ID and votes columns and pick "align-->right"
# Select the priority column and pick "align-->center"
# Select the first row and click on the "bold" button
# Grab the lines between the column headers to adjust the column widths
# Grab the sort bar on the far left (just above the "1" for row one)
# and move it down one row.  (Row one becomes a floating header)
# Voila!

# The team members
# FIXME: These names really should be external to this script
team = sorted('Steven Gary Greg Ken Jim Bill Jason Dirk Anatoly'.split())

# The elements to be picked out of the issue
PickList = [
    # sort key -- these are used to sort the entry
    'target_milestone', 'priority', 'votes_desc', 'creation_ts',
    # payload -- these are displayed
    'issue_id', 'votes', 'issue_type', 'target_milestone',
    'priority', 'assigned_to', 'short_desc',

# Conbert a leaf element into its value as a text string
# We assume it's "short enough" that there's only one substring
def Value(element):
    v = element.firstChild
    if v is None: return ''
    return v.nodeValue

# Parse the XML issues file and produce a DOM for it
import sys
if len(sys.argv) > 1: xml = sys.argv[1]
else: xml = 'issues.xml'
from xml.dom.minidom import parse
xml = parse(xml)

# Go through the issues in the DOM, pick out the elements we want,
# and put them in our list of issues.
issues = []
for issuezilla in xml.childNodes:
    # The Issuezilla element contains the issues
    if issuezilla.nodeType != issuezilla.ELEMENT_NODE: continue
    for issue in issuezilla.childNodes:
        # The issue elements contain the info for an issue
        if issue.nodeType != issue.ELEMENT_NODE: continue
        # Accumulate the pieces we want to include
        d = {}
        for element in issue.childNodes:
            if element.nodeName in PickList:
                d[element.nodeName] = Value(element)
        # convert 'votes' to numeric, ascending and descending
            v = int('0' + d['votes'])
        except KeyError:
            d['votes_desc'] = -v
            d['votes'] = v
        # Marshal the elements and add them to the list
        issues.append([ d[ix] for ix in PickList ])

# Transcribe the issues into comma-separated values.
# FIXME: parameterize the output file name
import csv
writer = csv.writer(open('editlist.csv', 'w'))
# header
writer.writerow(['ID', 'Votes', 'Type/Member', 'Milestone',
        'Pri', 'Owner', 'Summary/Comments'])
for issue in issues:
    row = issue[4:]        # strip off sort key
    #row[0] = """=hyperlink("http://scons.tigris.org/issues/show_bug.cgi?id=%s","%s")""" % (row[0],row[0])
    if row[3] == '-unspecified-': row[3] = 'triage'
    for member in team: writer.writerow(['','',member,'','','',''])

print "Exported %d issues to editlist.csv.  Ready to upload to Google."%len(issues)

# Local Variables:
# tab-width:4
# indent-tabs-mode:nil
# End:
# vim: set expandtab tabstop=4 shiftwidth=4: