[gitdm,1/1] Convert to Python 3

This is the result of running 2to3 and manually converting the comparison
functions to lambda expressions.

Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
---
 ConfigFile.py |   4 +-
 csvdump.py    |   4 +-
 database.py   |  22 +++++------
 gitdm         |  27 ++++++-------
 gitlog.py     |   8 ++--
 logparser.py  |   6 +--
 reports.py    | 106 +++++++++++++++-----------------------------------
 utils.py      |   4 +-
 8 files changed, 70 insertions(+), 111 deletions(-)

> This is the result of running 2to3 and manually converting the comparison
> functions to lambda expressions.
> 
> Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>

Hello Tom,

which is the branch that you use for the gitdm U-Boot stats? Is it 
master or u-boot? If it is u-boot, please, update gitlab to make u-boot 
the main branch.

The current patch applies to the master branch.

Best regards

Heinrich

> ---
>   ConfigFile.py |   4 +-
>   csvdump.py    |   4 +-
>   database.py   |  22 +++++------
>   gitdm         |  27 ++++++-------
>   gitlog.py     |   8 ++--
>   logparser.py  |   6 +--
>   reports.py    | 106 +++++++++++++++-----------------------------------
>   utils.py      |   4 +-
>   8 files changed, 70 insertions(+), 111 deletions(-)
> 
> diff --git a/ConfigFile.py b/ConfigFile.py
> index 3a1e208..e58c601 100644
> --- a/ConfigFile.py
> +++ b/ConfigFile.py
> @@ -147,10 +147,10 @@ def ReadFileType (filename):
>           m = regex_file_type.match (line)
>           if not m or len (m.groups ()) != 2:
>               ConfigFile.croak ('Funky file type line "%s"' % (line))
> -        if not patterns.has_key (m.group (1)):
> +        if m.group (1) not in patterns:
>               patterns[m.group (1)] = []
>           if m.group (1) not in order:
> -            print '%s not found, appended to the last order' % m.group (1)
> +            print('%s not found, appended to the last order' % m.group (1))
>               order.append (m.group (1))
>   
>           patterns[m.group (1)].append (re.compile (m.group (2), re.IGNORECASE))
> diff --git a/csvdump.py b/csvdump.py
> index 9d1a65e..55e38a0 100644
> --- a/csvdump.py
> +++ b/csvdump.py
> @@ -50,7 +50,7 @@ def store_patch(patch):
>           ChangeSets.append([patch.commit, str(patch.date),
>                              patch.email, domain, author, employer,
>                              patch.added, patch.removed])
> -        for (filetype, (added, removed)) in patch.filetypes.iteritems():
> +        for (filetype, (added, removed)) in patch.filetypes.items():
>               FileTypes.append([patch.commit, filetype, added, removed])
>   
>   
> @@ -82,7 +82,7 @@ def OutputCSV (file):
>       writer = csv.writer (file, quoting=csv.QUOTE_NONNUMERIC)
>       writer.writerow (['Name', 'Email', 'Affliation', 'Date',
>                         'Added', 'Removed', 'Changesets'])
> -    for date, stat in PeriodCommitHash.items():
> +    for date, stat in list(PeriodCommitHash.items()):
>           # sanitise names " is common and \" sometimes too
>           empl_name = stat.employer.name.replace ('"', '.').replace ('\\', '.')
>           author_name = stat.name.replace ('"', '.').replace ('\\', '.')
> diff --git a/database.py b/database.py
> index bf13227..b267ed5 100644
> --- a/database.py
> +++ b/database.py
> @@ -40,7 +40,7 @@ class Hacker:
>                   for edate, empl in self.employer[i]:
>                       if edate > date:
>                           return empl
> -        print 'OOPS.  ', self.name, self.employer, self.email, email, date
> +        print('OOPS.  ', self.name, self.employer, self.email, email, date)
>           return None # Should not happen
>   
>       def addpatch (self, patch):
> @@ -124,11 +124,11 @@ def LookupStoreHacker(name, email, mapunknown = True):
>   
>   
>   def AllHackers ():
> -    return HackersByID.values ()
> +    return list(HackersByID.values ())
>   
>   def DumpDB ():
>       out = open ('database.dump', 'w')
> -    names = HackersByName.keys ()
> +    names = list(HackersByName.keys ())
>       names.sort ()
>       for name in names:
>           h = HackersByName[name]
> @@ -149,7 +149,7 @@ def DumpDB ():
>   # push it backward through the changes we've already seen.
>   #
>   def ApplyFirstTag (tag):
> -    for n in HackersByName.keys ():
> +    for n in list(HackersByName.keys ()):
>           if HackersByName[n].versions:
>               HackersByName[n].versions = [tag]
>   
> @@ -185,7 +185,7 @@ def GetEmployer (name):
>           return e
>   
>   def AllEmployers ():
> -    return Employers.values ()
> +    return list(Employers.values ())
>   
>   #
>   # Certain obnoxious developers, who will remain nameless (because we
> @@ -215,8 +215,8 @@ class VirtualEmployer (Employer):
>           self.__init__ (name) # Reset counts just in case
>   
>       def store (self):
> -        if Employers.has_key (self.name):
> -            print Employers[self.name]
> +        if self.name in Employers:
> +            print(Employers[self.name])
>               sys.stderr.write ('WARNING: Virtual empl %s overwrites another\n'
>                                 % (self.name))
>           if len (self.splits) == 0:
> @@ -235,7 +235,7 @@ class FileType:
>           order = order or self.order
>   
>           for file_type in order:
> -            if patterns.has_key (file_type):
> +            if file_type in patterns:
>                   for patt in patterns[file_type]:
>                       if patt.search (filename):
>                           return file_type
> @@ -261,7 +261,7 @@ def MixVirtuals ():
>   EmailAliases = { }
>   
>   def AddEmailAlias (variant, canonical):
> -    if EmailAliases.has_key (variant):
> +    if variant in EmailAliases:
>           sys.stderr.write ('Duplicate email alias for %s\n' % (variant))
>       EmailAliases[variant] = canonical
>   
> @@ -288,7 +288,7 @@ def AddEmailEmployerMapping (email, employer, end = nextyear):
>           for i in range (0, len(l)):
>               date, xempl = l[i]
>               if date == end:  # probably both nextyear
> -                print 'WARNING: duplicate email/empl for %s' % (email)
> +                print('WARNING: duplicate email/empl for %s' % (email))
>               if date > end:
>                   l.insert (i, (end, empl))
>                   return
> @@ -305,7 +305,7 @@ def MapToEmployer (email, unknown = 0):
>           pass
>       namedom = email.split ('@')
>       if len (namedom) < 2:
> -        print 'Oops...funky email %s' % email
> +        print('Oops...funky email %s' % email)
>           return [(nextyear, GetEmployer ('Funky'))]
>       s = namedom[1].split ('.')
>       for dots in range (len (s) - 2, -1, -1):
> diff --git a/gitdm b/gitdm
> index 61318ad..f426cc7 100755
> --- a/gitdm
> +++ b/gitdm
> @@ -1,4 +1,4 @@
> -#!/usr/bin/pypy
> +#!/usr/bin/python3
>   #-*- coding:utf-8 -*-
>   #
>   
> @@ -15,7 +15,8 @@
>   
>   import database, csvdump, ConfigFile, reports
>   import getopt, datetime
> -import os, re, sys, rfc822, string, os.path
> +from email.utils import parsedate_tz
> +import os, re, sys, string, os.path
>   import logparser
>   from patterns import patterns
>   
> @@ -108,7 +109,7 @@ def ParseOpts():
>           elif opt[0] == '-p':
>               CSVPrefix = opt[1]
>           elif opt[0] == '-r':
> -            print 'Filter on "%s"' % (opt[1])
> +            print('Filter on "%s"' % (opt[1]))
>               FileFilter = re.compile(opt[1])
>           elif opt[0] == '-s':
>               AuthorSOBs = 0
> @@ -120,7 +121,7 @@ def ParseOpts():
>               ReportUnknowns = True
>           elif opt[0] == '-x':
>               CSVFile = open(opt[1], 'w')
> -            print "open output file " + opt[1] + "\n"
> +            print("open output file " + opt[1] + "\n")
>           elif opt [0] == '-w':
>               Aggregate = 'week'
>           elif opt [0] == '-y':
> @@ -172,7 +173,7 @@ DateMap = { }
>   
>   def AddDateLines(date, lines):
>       if lines > 1000000:
> -        print 'Skip big patch (%d)' % lines
> +        print('Skip big patch (%d)' % lines)
>           return
>       try:
>           DateMap[date] += lines
> @@ -180,7 +181,7 @@ def AddDateLines(date, lines):
>           DateMap[date] = lines
>   
>   def PrintDateStats():
> -    dates = DateMap.keys()
> +    dates = list(DateMap.keys())
>       dates.sort()
>       total = 0
>       datef = open('datelc.csv', 'w')
> @@ -195,7 +196,7 @@ def PrintDateStats():
>   # Let's slowly try to move some smarts into this class.
>   #
>   class patch:
> -    (ADDED, REMOVED) = range(2)
> +    (ADDED, REMOVED) = list(range(2))
>   
>       def __init__(self, commit):
>           self.commit = commit
> @@ -219,7 +220,7 @@ class patch:
>           self.reports.append(reporter)
>   
>       def addfiletype(self, filetype, added, removed):
> -        if self.filetypes.has_key(filetype):
> +        if filetype in self.filetypes:
>               self.filetypes[filetype][self.ADDED] += added
>               self.filetypes[filetype][self.REMOVED] += removed
>           else:
> @@ -330,7 +331,7 @@ def grabpatch(logpatch):
>           #
>           m = patterns['date'].match(Line)
>           if m:
> -            dt = rfc822.parsedate(m.group(2))
> +            dt = parsedate_tz(m.group(2))
>               p.date = datetime.date(dt[0], dt[1], dt[2])
>               if p.date > Today:
>                   sys.stderr.write('Funky date: %s\n' % p.date)
> @@ -389,7 +390,7 @@ def GripeAboutAuthorName(name):
>       if name in GripedAuthorNames:
>           return
>       GripedAuthorNames.append(name)
> -    print '%s is an author name, probably not what you want' % (name)
> +    print('%s is an author name, probably not what you want' % (name))
>   
>   def ApplyFileFilter(line, ignore):
>       #
> @@ -462,14 +463,14 @@ TotalChanged = TotalAdded = TotalRemoved = 0
>   #
>   # Snarf changesets.
>   #
> -print >> sys.stderr, 'Grabbing changesets...\r',
> +print('Grabbing changesets...\r', end=' ', file=sys.stderr)
>   
>   patches = logparser.LogPatchSplitter(sys.stdin)
>   printcount = CSCount = 0
>   
>   for logpatch in patches:
>       if (printcount % 50) == 0:
> -        print >> sys.stderr, 'Grabbing changesets...%d\r' % printcount,
> +        print('Grabbing changesets...%d\r' % printcount, end=' ', file=sys.stderr)
>       printcount += 1
>   
>       # We want to ignore commits on svn tags since in Subversion
> @@ -528,7 +529,7 @@ for logpatch in patches:
>           CSCount += 1
>       csvdump.AccumulatePatch(p, Aggregate)
>       csvdump.store_patch(p)
> -print >> sys.stderr, 'Grabbing changesets...done       '
> +print('Grabbing changesets...done       ', file=sys.stderr)
>   
>   if DumpDB:
>       database.DumpDB()
> diff --git a/gitlog.py b/gitlog.py
> index 71efee1..4b1c5d6 100644
> --- a/gitlog.py
> +++ b/gitlog.py
> @@ -61,7 +61,7 @@ S_DONE = 5
>   def get_header(patch, line, input):
>       if line == '':
>           if patch.author == '':
> -            print 'Funky auth line in', patch.commit
> +            print('Funky auth line in', patch.commit)
>               patch.author = database.LookupStoreHacker('Unknown',
>                                                         'unknown@hacker.net')
>           return S_DESC
> @@ -78,7 +78,7 @@ def get_header(patch, line, input):
>   
>   def get_desc(patch, line, input):
>       if not line:
> -        print 'Missing desc in', patch.commit
> +        print('Missing desc in', patch.commit)
>           return S_CHANGELOG
>       patch.desc = line
>       line = getline(input)
> @@ -188,7 +188,7 @@ def grabpatch(input):
>           return None
>       m = patterns['commit'].match(line)
>       if not m:
> -        print 'noncommit', line
> +        print('noncommit', line)
>           return None
>       p = patch(m.group(1))
>       state = S_HEADER
> @@ -199,7 +199,7 @@ def grabpatch(input):
>           line = getline(input)
>           if line is None:
>               if state != S_NUMSTAT:
> -                print 'Ran out of patch', state
> +                print('Ran out of patch', state)
>                   return None
>               return p
>           state = grabbers[state](p, line, input)
> diff --git a/logparser.py b/logparser.py
> index b375034..88293c5 100644
> --- a/logparser.py
> +++ b/logparser.py
> @@ -41,7 +41,7 @@ class LogPatchSplitter:
>       def __iter__(self):
>           return self
>   
> -    def next(self):
> +    def __next__(self):
>           patch = self.__grab_patch__()
>           if not patch:
>               raise StopIteration
> @@ -85,6 +85,6 @@ if __name__ == '__main__':
>       patches = LogPatchSplitter(sys.stdin)
>   
>       for patch in patches:
> -        print '---------- NEW PATCH ----------'
> +        print('---------- NEW PATCH ----------')
>           for line in patch:
> -            print line,
> +            print(line, end=' ')
> diff --git a/reports.py b/reports.py
> index d7a96bc..3e03e69 100644
> --- a/reports.py
> +++ b/reports.py
> @@ -69,11 +69,8 @@ def EndReport():
>   #
>   # Comparison and report generation functions.
>   #
> -def ComparePCount(h1, h2):
> -    return len(h2.patches) - len(h1.patches)
> -
>   def ReportByPCount(hlist, cscount):
> -    hlist.sort(ComparePCount)
> +    hlist.sort(key=lambda h: -len(h.patches))
>       count = 0
>       BeginReport('Developers with the most changesets')
>       for h in hlist:
> @@ -87,11 +84,8 @@ def ReportByPCount(hlist, cscount):
>               break
>       EndReport()
>               
> -def CompareLChanged(h1, h2):
> -    return h2.changed - h1.changed
> -
>   def ReportByLChanged(hlist, totalchanged):
> -    hlist.sort(CompareLChanged)
> +    hlist.sort(key=lambda h: -h.changed)
>       count = 0
>       BeginReport('Developers with the most changed lines')
>       for h in hlist:
> @@ -103,11 +97,8 @@ def ReportByLChanged(hlist, totalchanged):
>               break
>       EndReport()
>               
> -def CompareLRemoved(h1, h2):
> -    return (h2.removed - h2.added) - (h1.removed - h1.added)
> -
>   def ReportByLRemoved(hlist, totalremoved):
> -    hlist.sort(CompareLRemoved)
> +    hlist.sort(key=lambda h: h.added - h.removed)
>       count = 0
>       BeginReport('Developers with the most lines removed')
>       for h in hlist:
> @@ -121,11 +112,8 @@ def ReportByLRemoved(hlist, totalremoved):
>               break
>       EndReport()
>   
> -def CompareEPCount(e1, e2):
> -    return e2.count - e1.count
> -
>   def ReportByPCEmpl(elist, cscount):
> -    elist.sort(CompareEPCount)
> +    elist.sort(key=lambda e: -e.count)
>       count = 0
>       BeginReport('Top changeset contributors by employer')
>       for e in elist:
> @@ -137,11 +125,8 @@ def ReportByPCEmpl(elist, cscount):
>       EndReport()
>   
>   
> -def CompareELChanged(e1, e2):
> -    return e2.changed - e1.changed
> -
>   def ReportByELChanged(elist, totalchanged):
> -    elist.sort(CompareELChanged)
> +    elist.sort(key=lambda e: -e.changed)
>       count = 0
>       BeginReport('Top lines changed by employer')
>       for e in elist:
> @@ -154,11 +139,8 @@ def ReportByELChanged(elist, totalchanged):
>   
>   
>   
> -def CompareSOBs(h1, h2):
> -    return len(h2.signoffs) - len(h1.signoffs)
> -
>   def ReportBySOBs(hlist):
> -    hlist.sort(CompareSOBs)
> +    hlist.sort(key = lambda h: -len(h.signoffs))
>       totalsobs = 0
>       for h in hlist:
>           totalsobs += len(h.signoffs)
> @@ -176,11 +158,8 @@ def ReportBySOBs(hlist):
>   #
>   # Reviewer reporting.
>   #
> -def CompareRevs(h1, h2):
> -    return len(h2.reviews) - len(h1.reviews)
> -
>   def ReportByRevs(hlist):
> -    hlist.sort(CompareRevs)
> +    hlist.sort(key=lambda h: -len(h.reviews))
>       totalrevs = 0
>       for h in hlist:
>           totalrevs += len(h.reviews)
> @@ -198,11 +177,8 @@ def ReportByRevs(hlist):
>   #
>   # tester reporting.
>   #
> -def CompareTests(h1, h2):
> -    return len(h2.tested) - len(h1.tested)
> -
>   def ReportByTests(hlist):
> -    hlist.sort(CompareTests)
> +    hlist.sort(key=lambda h: -len(h.tested))
>       totaltests = 0
>       for h in hlist:
>           totaltests += len(h.tested)
> @@ -217,11 +193,8 @@ def ReportByTests(hlist):
>               break
>       EndReport()
>   
> -def CompareTestCred(h1, h2):
> -    return h2.testcred - h1.testcred
> -
>   def ReportByTestCreds(hlist):
> -    hlist.sort(CompareTestCred)
> +    hlist.sort(key=lambda h: -h.testcred)
>       totaltests = 0
>       for h in hlist:
>           totaltests += h.testcred
> @@ -240,11 +213,8 @@ def ReportByTestCreds(hlist):
>   #
>   # Reporter reporting.
>   #
> -def CompareReports(h1, h2):
> -    return len(h2.reports) - len(h1.reports)
> -
>   def ReportByReports(hlist):
> -    hlist.sort(CompareReports)
> +    hlist.sort(key=lambda h: -len(h.reports))
>       totalreps = 0
>       for h in hlist:
>           totalreps += len(h.reports)
> @@ -259,11 +229,8 @@ def ReportByReports(hlist):
>               break
>       EndReport()
>   
> -def CompareRepCred(h1, h2):
> -    return h2.repcred - h1.repcred
> -
>   def ReportByRepCreds(hlist):
> -    hlist.sort(CompareRepCred)
> +    hlist.sort(key=lambda h: -h.repcred)
>       totalreps = 0
>       for h in hlist:
>           totalreps += h.repcred
> @@ -280,14 +247,11 @@ def ReportByRepCreds(hlist):
>   #
>   # Versions.
>   #
> -def CompareVersionCounts(h1, h2):
> -    if h1.versions and h2.versions:
> -        return len(h2.versions) - len(h1.versions)
> -    if h2.versions:
> -        return 1
> -    if h1.versions:
> -        return -1
> -    return 0
> +def VersionCount(h):
> +    if h.versions:
> +        return len(h.versions)
> +    else:
> +        return 0
>   
>   def MissedVersions(hv, allv):
>       missed = [v for v in allv if v not in hv]
> @@ -295,7 +259,7 @@ def MissedVersions(hv, allv):
>       return ' '.join(missed)
>   
>   def ReportVersions(hlist):
> -    hlist.sort(CompareVersionCounts)
> +    hlist.sort(key=lambda h: -VersionCount(h))
>       BeginReport('Developers represented in the most kernel versions')
>       count = 0
>       allversions = hlist[0].versions
> @@ -307,11 +271,8 @@ def ReportVersions(hlist):
>       EndReport()
>   
>   
> -def CompareESOBs(e1, e2):
> -    return e2.sobs - e1.sobs
> -
>   def ReportByESOBs(elist):
> -    elist.sort(CompareESOBs)
> +    elist.sort(key=lambda e: -e.sobs)
>       totalsobs = 0
>       for e in elist:
>           totalsobs += e.sobs
> @@ -325,11 +286,8 @@ def ReportByESOBs(elist):
>               break
>       EndReport()
>      
> -def CompareHackers(e1, e2):
> -    return len(e2.hackers) - len(e1.hackers)
> -
>   def ReportByEHackers(elist):
> -    elist.sort(CompareHackers)
> +    elist.sort(key=lambda e: -len(e.hackers))
>       totalhackers = 0
>       for e in elist:
>           totalhackers += len(e.hackers)
> @@ -375,7 +333,7 @@ def ReportUnknowns(hlist, cscount):
>       # mapping to (Unknown) is happening or not.
>       #
>       ulist = [ h for h in hlist if IsUnknown(h) ]
> -    ulist.sort(ComparePCount)
> +    ulist.sort(lambda h: len(h.patches))
>       count = 0
>       BeginReport('Developers with unknown affiliation')
>       for h in ulist:
> @@ -398,46 +356,46 @@ def ReportByFileType(hacker_list):
>           by_hacker = {}
>           for patch in h.patches:
>               # Get a summary by hacker
> -            for (filetype, (added, removed)) in patch.filetypes.iteritems():
> -                if by_hacker.has_key(filetype):
> +            for (filetype, (added, removed)) in patch.filetypes.items():
> +                if filetype in by_hacker:
>                       by_hacker[filetype][patch.ADDED] += added
>                       by_hacker[filetype][patch.REMOVED] += removed
>                   else:
>                       by_hacker[filetype] = [added, removed]
>   
>                   # Update the totals
> -                if total.has_key(filetype):
> +                if filetype in total:
>                       total[filetype][patch.ADDED] += added
>                       total[filetype][patch.REMOVED] += removed
>                   else:
>                       total[filetype] = [added, removed, []]
>   
>           # Print a summary by hacker
> -        print h.name
> -        for filetype, counters in by_hacker.iteritems():
> -            print '\t', filetype, counters
> +        print(h.name)
> +        for filetype, counters in by_hacker.items():
> +            print('\t', filetype, counters)
>               h_added = by_hacker[filetype][patch.ADDED]
>               h_removed = by_hacker[filetype][patch.REMOVED]
>               total[filetype][2].append([h.name, h_added, h_removed])
>   
>       # Print the global summary
>       BeginReport('Contributions by type and developers')
> -    for filetype, (added, removed, hackers) in total.iteritems():
> -        print filetype, added, removed
> +    for filetype, (added, removed, hackers) in total.items():
> +        print(filetype, added, removed)
>           for h, h_added, h_removed in hackers:
> -            print '\t%s: [%d, %d]' % (h, h_added, h_removed)
> +            print('\t%s: [%d, %d]' % (h, h_added, h_removed))
>   
>       # Print the very global summary
>       BeginReport('General contributions by type')
> -    for filetype, (added, removed, hackers) in total.iteritems():
> -        print filetype, added, removed
> +    for filetype, (added, removed, hackers) in total.items():
> +        print(filetype, added, removed)
>   
>   #
>   # The file access report is a special beast.
>   #
>   def FileAccessReport(name, accesses, total):
>       outf = open(name, 'w')
> -    files = accesses.keys()
> +    files = list(accesses.keys())
>       files.sort()
>       for file in files:
>           a = accesses[file]
> diff --git a/utils.py b/utils.py
> index 2b3be5d..9f17911 100644
> --- a/utils.py
> +++ b/utils.py
> @@ -21,7 +21,7 @@ class accumulator:
>               return default
>   
>       def append(self, key, item, unique = False):
> -        if unique and self._data.has_key(key) and \
> +        if unique and key in self._data and \
>              item in self._data[key]:
>               return
>           try:
> @@ -30,7 +30,7 @@ class accumulator:
>               self._data[key] = [item]
>   
>       def keys(self):
> -        return self._data.keys()
> +        return list(self._data.keys())
>   
>       def __getitem__(self, key):
>           return self._data[key]

On Wed, Jun 07, 2023 at 07:57:54AM +0200, Heinrich Schuchardt wrote:

> This is the result of running 2to3 and manually converting the comparison
> functions to lambda expressions.
> 
> Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
> ---
>  ConfigFile.py |   4 +-
>  csvdump.py    |   4 +-
>  database.py   |  22 +++++------
>  gitdm         |  27 ++++++-------
>  gitlog.py     |   8 ++--
>  logparser.py  |   6 +--
>  reports.py    | 106 +++++++++++++++-----------------------------------
>  utils.py      |   4 +-
>  8 files changed, 70 insertions(+), 111 deletions(-)
> 

Is this really versus our current tree? It's already running in python3.
If there's specific "this is better python" changes, we should send them
to Jon first and sync back.