def collect(ip,arg):
""" collect foo/a.txt rec:bar=*.py
Copies foo/a.txt to ~/_ipython/collect/foo/a.txt and *.py from bar,
likewise
Without args, try to open ~/_ipython/collect dir (in win32 at least).
"""
from IPython.external.path import path
basedir = path(ip.ipython_dir + '/collect')
try:
fs = mglob.expand(arg.split(None,1)[1])
except IndexError:
os.startfile(basedir)
return
for f in fs:
f = path(f)
trg = basedir / f.splitdrive()[1].lstrip('/\\')
if f.isdir():
print "mkdir",trg
trg.makedirs()
continue
dname = trg.dirname()
if not dname.isdir():
dname.makedirs()
print f,"=>",trg
shutil.copy2(f,trg)
def collect(ip, arg):
""" collect foo/a.txt rec:bar=*.py
Copies foo/a.txt to ~/_ipython/collect/foo/a.txt and *.py from bar,
likewise
Without args, try to open ~/_ipython/collect dir (in win32 at least).
"""
from IPython.external.path import path
basedir = path(ip.options.IPYTHONDIR + '/collect')
try:
fs = mglob.expand(arg.split(None, 1)[1])
except IndexError:
os.startfile(basedir)
return
for f in fs:
f = path(f)
trg = basedir / f.splitdrive()[1].lstrip('/\\')
if f.isdir():
print "mkdir", trg
trg.makedirs()
continue
dname = trg.dirname()
if not dname.isdir():
dname.makedirs()
print f, "=>", trg
shutil.copy2(f, trg)
def sniff_repository_type(_path):
if (path(_path) / '.hg').exists():
return 'hg'
if (path(_path) / '.git').exists():
return 'git'
if (path(_path) / '.svn').exists():
return 'svn'
raise NotImplementedError("unabled to determine repository type")
def get_paths(self):
try:
return self.__paths
except AttributeError:
self.__paths = [path(p)
for p in self.split('\n') if os.path.exists(p)]
return self.__paths
def get_paths(self):
try:
return self.__paths
except AttributeError:
self.__paths = [
path(p) for p in self.split('\n') if os.path.exists(p)
]
return self.__paths
def __getattr__(self,name):
if name in self.ents:
if self.path.endswith('/'):
sep = ''
else:
sep = '/'
tgt = self.path + sep + pathobj_unmangle(name)
#print "tgt",tgt
if os.path.isdir(tgt):
return PathObj(tgt)
if os.path.isfile(tgt):
return path(tgt)
raise AttributeError, name # <<< DON'T FORGET THIS LINE !!
def __getattr__(self, name):
if name in self.ents:
if self.path.endswith('/'):
sep = ''
else:
sep = '/'
tgt = self.path + sep + pathobj_unmangle(name)
#print "tgt",tgt
if os.path.isdir(tgt):
return PathObj(tgt)
if os.path.isfile(tgt):
return path(tgt)
raise AttributeError, name # <<< DON'T FORGET THIS LINE !!
def main():
all = path('.').files('test_*py')
results = {}
res_exc = [None]
def exchook(self,*e):
res_exc[0] = [e]
ip.IP.set_custom_exc((Exception,), exchook)
startdir = os.getcwd()
for test in all:
print test
res_exc[0] = 'ok'
os.chdir(startdir)
ip.runlines(test.text())
results[str(test)] = res_exc[0]
os.chdir(startdir)
pprint.pprint(results)
def main():
all = path('.').files('test_*py')
results = {}
res_exc = [None]
def exchook(self, *e):
res_exc[0] = [e]
ip.IP.set_custom_exc((Exception, ), exchook)
startdir = os.getcwd()
for test in all:
print test
res_exc[0] = 'ok'
os.chdir(startdir)
ip.runlines(test.text())
results[str(test)] = res_exc[0]
os.chdir(startdir)
pprint.pprint(results)
characters can be found in our source tree.
"""
# Config
# If true, all lines that have tabs are printed, with line number
full_report_tabs = True
# If true, all lines that have tabs are printed, with line number
full_report_rets = False
# Code begins
from IPython.external.path import path
rets = []
tabs = []
for f in path("..").walkfiles("*.py"):
errs = ""
cont = f.bytes()
if "\t" in cont:
errs += "t"
tabs.append(f)
if "\r" in cont:
errs += "r"
rets.append(f)
if errs:
print("%3s" % errs, f)
if "t" in errs and full_report_tabs:
for ln, line in enumerate(f.lines()):
from IPython.external.path import path
fs = path('..').walkfiles('*.py')
for f in fs:
errs = ''
cont = f.bytes()
if '\t' in cont:
errs+='t'
if '\r' in cont:
errs+='r'
if errs:
print "%3s" % errs, f
from IPython.external.path import path
fs = path('..').walkfiles('*.py')
for f in fs:
errs = ''
cont = f.bytes()
if '\t' in cont:
errs += 't'
if '\r' in cont:
errs += 'r'
if errs:
print "%3s" % errs, f
characters can be found in our source tree.
"""
# Config
# If true, all lines that have tabs are printed, with line number
full_report_tabs = True
# If true, all lines that have tabs are printed, with line number
full_report_rets = False
# Code begins
from IPython.external.path import path
rets = []
tabs = []
for f in path('..').walkfiles('*.py'):
errs = ''
cont = f.bytes()
if '\t' in cont:
errs += 't'
tabs.append(f)
if '\r' in cont:
errs += 'r'
rets.append(f)
if errs:
print "%3s" % errs, f
if 't' in errs and full_report_tabs:
for ln, line in enumerate(f.lines()):
def upgrade_dir(srcdir, tgtdir):
""" Copy over all files in srcdir to tgtdir w/ native line endings
Creates .upgrade_report in tgtdir that stores md5sums of all files
to notice changed files b/w upgrades.
"""
def pr(s):
print s
junk = ['.svn','ipythonrc*','*.pyc', '*.pyo', '*~', '.hg']
def ignorable(p):
for pat in junk:
if p.startswith(pat) or p.fnmatch(pat):
return True
return False
modded = []
files = [path(srcdir).relpathto(p) for p in path(srcdir).walkfiles()]
#print files
rep = tgtdir / '.upgrade_report'
try:
rpt = pickle.load(rep.open())
except:
rpt = {}
for f in files:
if ignorable(f):
continue
src = srcdir / f
tgt = tgtdir / f
if not tgt.isfile():
pr("Creating %s" % str(tgt))
tgt.write_text(src.text())
rpt[str(tgt)] = md5.new(tgt.text()).hexdigest()
else:
cont = tgt.text()
sum = rpt.get(str(tgt), None)
#print sum
if sum and md5.new(cont).hexdigest() == sum:
pr("%s: Unedited, installing new version" % tgt)
tgt.write_text(src.text())
rpt[str(tgt)] = md5.new(tgt.text()).hexdigest()
else:
pr(' == Modified, skipping %s, diffs below == ' % tgt)
#rpt[str(tgt)] = md5.new(tgt.bytes()).hexdigest()
real = showdiff(tgt,src)
pr('') # empty line
if not real:
pr("(Ok, it was identical, only upgrading checksum)")
rpt[str(tgt)] = md5.new(tgt.text()).hexdigest()
else:
modded.append(tgt)
#print rpt
pickle.dump(rpt, rep.open('w'))
if modded:
print "\n\nDelete the following files manually (and rerun %upgrade)\nif you need a full upgrade:"
for m in modded:
print m
else:
cont = tgt.text()
sum = rpt.get(str(tgt), None)
#print sum
if sum and md5.new(cont).hexdigest() == sum:
pr("%s: Unedited, installing new version" % tgt)
tgt.write_text(src.text())
rpt[str(tgt)] = md5.new(tgt.text()).hexdigest()
else:
pr(' == Modified, skipping %s, diffs below == ' % tgt)
#rpt[str(tgt)] = md5.new(tgt.bytes()).hexdigest()
real = showdiff(tgt,src)
pr('') # empty line
if not real:
pr("(Ok, it was identical, only upgrading checksum)")
rpt[str(tgt)] = md5.new(tgt.text()).hexdigest()
else:
modded.append(tgt)
#print rpt
pickle.dump(rpt, rep.open('w'))
if modded:
print "\n\nDelete the following files manually (and rerun %upgrade)\nif you need a full upgrade:"
for m in modded:
print m
import sys
if __name__ == "__main__":
upgrade_dir(path(sys.argv[1]), path(sys.argv[2]))
'''
exit(1)
print "Search path: %s" % str(sys.argv[1])
remove = False
link = False
if len(sys.argv) == 3:
if sys.argv[2] == 'remove':
remove = True
elif sys.argv[2] == 'symlink':
link = True
p = path.path(sys.argv[1])
d = {}
c = {}
dup = set()
for f in p.walk():
if not f.isdir() and not f.islink():
s = io.open(f, 'rb')
try:
h = md5.md5(s.read())
h = h.hexdigest()
#print '{}'.format(h)
if h not in d:
d[h] = []
d[h].append(f)
c[h] = len(d[h])
if c[h] > 1:
characters can be found in our source tree.
"""
# Config
# If true, all lines that have tabs are printed, with line number
full_report_tabs = True
# If true, all lines that have tabs are printed, with line number
full_report_rets = False
# Code begins
from IPython.external.path import path
rets = []
tabs = []
for f in path('..').walkfiles('*.py'):
errs = ''
cont = f.bytes()
if '\t' in cont:
errs+='t'
tabs.append(f)
if '\r' in cont:
errs+='r'
rets.append(f)
if errs:
print "%3s" % errs, f
if 't' in errs and full_report_tabs:
for ln,line in enumerate(f.lines()):
import pandas
import datetime
import calendar
from pandas import DataFrame, Index, Series
from IPython.external.path import path
datapath = path('~/workspace/math/facts').expanduser()
def get_gas_prices():
# http://www.eia.gov/dnav/pet/hist/LeafHandler.ashx?n=PET&s=EMA_EPM0_PTA_NUS_DPG&f=M
# http://www.eia.gov/dnav/pet/hist_xls/EMA_EPM0_PTA_NUS_DPGm.xls
gas_cpg = 'EMA_EPM0_PTA_NUS_DPGm.xls'
xls = pandas.ExcelFile(datapath / gas_cpg)
gasdata = xls.parse(xls.sheet_names[1], skiprows=(0, 1))
gasdata['prices'] = gasdata[gasdata.columns[1]]
del gasdata[gasdata.columns[1]]
gasdata['date'] = gasdata['Date']
del gasdata['Date']
gasdata.index = gasdata['date']
return gasdata.resample('M', convention='end')
def clean_cpiai(path):
def normalize(path):
with open(path) as f:
yield ('date', 'price')
for i, l in enumerate(f):
l = l.strip()
if i > 17:
l = l.strip()
#!/usr/bin/env python
# encoding: utf-8
from __future__ import print_function
"""
progname
"""
import BeautifulSoup
from IPython.external.path import path
import logging
DATAPATH = path('~/workspace/math/facts/data').expanduser()
def parse_tax_rates_table():
# https://en.wikipedia.org/wiki/Income_tax_in_the_United_States
datapath = DATAPATH / 'taxes'
htmlpath = datapath / 'Income tax in the United States - Wikipedia, the free encyclopedia.html'
bs = BeautifulSoup.BeautifulSoup(open(htmlpath))
def find_unadjusted_table(bs):
# TODO
tables = bs.findAll('table')
#print(len(tables))
return tables[-2]
table = find_unadjusted_table(bs)
row = table.findAll('tr')[1]
headings = tuple(col.text for col in row.findAll('th'))
print(headings)
