commit f2e836da0dafde74e678a1943dc281dd6669eaae
Author: Erik Osheim <erik@osheim.org>
Date:   Mon Aug 2 11:14:18 2010 -0400

    initial import

diff --git a/cleanse b/cleanse
new file mode 100755
index 0000000..629a41e
--- /dev/null
+++ b/cleanse
@@ -0,0 +1,22 @@
+#!/usr/bin/python
+#
+# by Erik osheim
+import optparse, os
+
+epilog   = 'Remove various junky files left lying around.'
+patterns = ('*~', '.#*', '#*', '*.orig', '*-')
+
+parser = optparse.OptionParser(epilog=epilog)
+parser.add_option('-f', '--fake', dest='fake', action='store_true',
+                  help="print files; do not remove them")
+opts, args = parser.parse_args()
+
+stanzas = ' -o '.join(["-name '%s'" % x for x in patterns])
+cmd     = "find . -type f \\( %s \\) -print" % stanzas
+
+if opts.fake:
+    print "files that would be deleted:"
+else:
+    cmd += " -exec rm {} ';'";
+
+os.system(cmd)
diff --git a/histo b/histo
new file mode 100755
index 0000000..5f27a1a
--- /dev/null
+++ b/histo
@@ -0,0 +1,105 @@
+#!/usr/bin/python
+import optparse
+import os
+import sys
+
+# this is a dictionary of all the "items" (lines) that have been seen, mapping
+# to a count of how many times it was seen.
+seen   = {}
+
+parser = optparse.OptionParser()
+parser.set_defaults(draw=False, nullok=False, valsort=False, reverse=False,
+                    hwidth=None, percent=False, strip=False, delim=' ',
+                    swap=False, josh=False, nopad=False)
+parser.add_option('-d', '--draw', action='store_true', dest='draw',
+                  help='draw histogram with horizontal graph')
+parser.add_option('-D', '--delim', action='store', type="string", dest='delim',
+                  help='use STR as delimiter (default: " ")', metavar='STR')
+parser.add_option('-f', '--frequency', action='store_true', dest='valsort',
+                  help='sort by frequency')
+parser.add_option('-J', '--josh', action='store_true', dest='josh',
+                  help='josh rosenberg compatibility mode')
+parser.add_option('-N', '--nopad', action='store_true', dest='nopad',
+                  help='do not column-align output')
+parser.add_option('-p', '--percent', action='store_true', dest='percent',
+                  help='show percent rather than count')
+parser.add_option('-r', '--reverse', action='store_true', dest='reverse',
+                  help='reverse sort')
+parser.add_option('-s', '--strip', action='store_true', dest='strip',
+                  help='strip input lines')
+parser.add_option('-S', '--swap', action='store_true', dest='swap',
+                  help='swap output fields')
+parser.add_option('-w', '--width', action='store', type='int', dest='hwidth',
+                  help='use NUM characters for graph drawing', metavar="NUM")
+parser.add_option('-z', '--null', action='store_true', dest='nullok',
+                  help='allow null lines')
+(opts, args) = parser.parse_args()
+
+# allow some special delimiters to be translated
+delims = {'\\t': '\t', '\\v': '\v', '\\0': '\0'}
+opts.delim = delims.get(opts.delim, opts.delim)
+
+# josh rosenberg compatiblity mode
+if opts.josh:
+    opts.delim = '\t'
+    opts.swap  = True
+    opts.nopad = True
+
+# process each line on stdin. we can optionally strip whitespace, and we always
+# strip a trailing newline. by default we ignore empty lines, but nullok causes
+# us to count them as well.
+for line in sys.stdin:
+    key = line[:-1]
+    if opts.strip:
+        key = key.strip()
+    if key or opts.nullok:
+        seen.setdefault(key, 0)
+        seen[key] += 1
+
+# sort either by key or by count, depending on the user wanted
+keys = seen.keys()
+if opts.valsort:
+    cmpf = lambda a, b: cmp(seen[b], seen[a])
+else:
+    cmpf = lambda a, b: cmp(a, b)
+keys.sort(cmpf)
+
+# optionally reverse the sort if the user wanted
+if opts.reverse: keys.reverse()
+
+# determine max key length, max count, and total count; also, the printed length
+# of the max count.
+kmax, cmax, total = 0, 0, 0
+for key in keys:
+    cmax, kmax, total = max(seen[key], cmax), max(len(key), kmax), total + seen[key]
+smax = len(str(cmax))
+
+# if we need to draw, we should make some guesses about the screen size. we
+# should probably never try to draw more than 40 characters worth of stars.
+width  = os.environ.get("COLUMNS", 80)
+if opts.hwidth is None:
+    opts.hwidth = min(abs(width - kmax), 40)
+
+for key in keys:
+    # figure out how to draw the key
+    if opts.nopad:
+        vkey = key
+    else:
+        vkey = '%-*s' % (kmax, key)
+
+    # figure out how to draw the value
+    if opts.draw:
+        ratio = float(seen[key]) / cmax
+        nstar = int(round(opts.hwidth * ratio))
+        value = '%*s' % (opts.hwidth, '*' * nstar)
+    elif opts.percent:
+        perc = float(seen[key]) * 100 / total
+        value = '%5.1f%%' % perc
+    else:
+        value = '%*d' % (smax, seen[key])
+
+    # figure out which order to write them in
+    if opts.swap:
+        print "%s%s%s" % (vkey, opts.delim, value)
+    else:
+        print "%s%s%s" % (value, opts.delim, key)
diff --git a/patchinfo b/patchinfo
new file mode 100755
index 0000000..c06c15f
--- /dev/null
+++ b/patchinfo
@@ -0,0 +1,83 @@
+#!/usr/bin/python
+#
+# by Erik Osheim
+#
+# available under the GNU GPL version 2.
+#
+# This program is designed to provide analysis of patches or diff files.
+
+import optparse
+import re
+import sys
+
+# matches the diff stanza, which have no leading identifier
+diff_re = re.compile('^[0-9a-zA-Z,]+$')
+
+# analyze a particular patch, which is assumed to contain multiple files
+def info(f, verbose=False):
+    paths = []
+    l = 0
+    first = True
+    for line in f:
+        if first:
+            if diff_re.match(line):
+                paths.append(['-', 0, 0, 0, 0])
+            first = False
+
+        if diff_re.match(line):
+            if not paths:
+                paths.append(['-', 0, 0, 0, 0])
+            paths[-1][1] += 1
+        elif line.startswith('diff'):
+            path = line.split()[-1]
+            l = max(l, len(path))
+            paths.append([path, 0, 0, 0, 0])
+        elif line.startswith('Index'):
+            path = line.split()[1]
+            l = max(l, len(path))
+            paths.append([path, 0, 0, 0, 0])
+        elif line.startswith('---') or line.startswith('+++') or line.startswith('='):
+            pass
+        elif line.startswith('@@'):
+            paths[-1][1] += 1
+        elif line.startswith('+') or line.startswith('>'):
+            paths[-1][2] += 1
+            paths[-1][3] += 1
+        elif line.startswith('-') or line.startswith('<'):
+            paths[-1][2] -= 1
+            paths[-1][4] += 1
+
+    if len(paths) != 1:
+        title = '%d paths' % len(paths)
+    else:
+        title = '1 path'
+
+    if verbose:
+        title = 'TOTAL ' + title
+    else:
+        l = len(title)
+
+    totals = [0, 0, 0, 0]
+    for info in paths:
+        for i in range(0, 4):
+            totals[i] += info[i + 1]
+
+    lens = [len(str(n)) + 1 for n in totals[:2]]
+    fmt  = '%%-%ds %%%dd groups, %%+%dd lines (+%%d -%%d)' % tuple([l] + lens)
+    for info in paths:
+        if verbose:
+            print fmt % tuple(info)
+
+    print fmt % tuple([title] + totals)
+
+if __name__ == "__main__":
+    op = optparse.OptionParser()
+    op.add_option('-v', '--verbose', dest='verbose', action='store_true')
+
+    opts, args = op.parse_args()
+
+    if args:
+        for arg in args:
+            info(open(arg, 'r'), verbose=opts.verbose)
+    else:
+        info(sys.stdin, verbose=opts.verbose)
diff --git a/sgrep b/sgrep
new file mode 100755
index 0000000..a5e8be2
--- /dev/null
+++ b/sgrep
@@ -0,0 +1,11 @@
+#!/bin/sh
+grep --binary-files=without-match \
+    --exclude-dir=CVS \
+    --exclude-dir=.svn \
+    --exclude-dir=.hg \
+    --exclude-dir=.git \
+    --exclude '*~' \
+    --exclude '.#*' \
+    --exclude '#*' \
+    --exclude '*-' \
+    "$@"
diff --git a/unbox b/unbox
new file mode 100755
index 0000000..47c4286
--- /dev/null
+++ b/unbox
@@ -0,0 +1,78 @@
+#!/usr/bin/env python
+#
+# by Erik Osheim
+#
+# This program unboxes various archive 
+
+import os
+import shutil
+from subprocess import call
+import sys
+import tempfile
+
+# list each kind of archive in terms of extension, and then program and
+# arguments to use. Use None to indicate where the filename should be placed.
+handlers = {
+    '.zip':     ['unzip', None],
+    '.tgz':     ['tar', 'xvzf', None],
+    '.tbz':     ['tar', 'xvjf', None],
+    '.tar.gz':  ['tar', 'xvzf', None],
+    '.tar.bz2': ['tar', 'xvjf', None],
+    #'.tar.xz':  ['tar', 'xvJz', None],
+}
+
+# program usage
+def usage(err=None):
+    retval = 0
+    if err:
+        retval = 1
+        print 'error: %s' % err
+    print 'usage: unbox ARCHIVE'
+    sys.exit(retval)
+
+if __name__ == "__main__":
+    args = sys.argv[1:]
+    if not args:
+        usage()
+    elif len(args) > 1:
+        usage('too many arguments')
+    path = args[0]
+    if not os.path.exists(path):
+        usage('path %r does not exist' % path)
+
+    cmd = None
+    for ext in handlers:
+        if path.endswith(ext):
+            cmd  = handlers[ext]
+            base = path[:-len(ext)]
+    if not cmd:
+        usage("can't handle path %r" % path)
+
+    d = tempfile.mkdtemp(prefix='unbox-')
+
+    fd, path2 = tempfile.mkstemp(dir=d)
+    os.write(fd, open(path, 'rb').read())
+    os.close(fd)
+
+    cwd = os.getcwd()
+    print d
+    os.chdir(d)
+
+    base2 = os.path.basename(path2)
+
+    # put the path into the command
+    cmd = [x or base2 for x in cmd]
+
+    retcode = call(cmd)
+    if retcode != 0:
+        print 'error: %s failed (%d)' % (cmd[0], retcode)
+        sys.exit(1)
+
+    os.unlink(path2)
+
+    items = os.listdir(d)
+    if len(items) == 1:
+        shutil.move(d + '/' + items[0], cwd)
+        shutil.rmtree(d)
+    else:
+        shutil.move(d, cwd + '/' + base)