Remove resources/python. All that stuff belongs in dotfiles/lib/python

dotfiles/lib/python/birthday_problem.py

@@ -0,0 +1,73 @@
+import random
+import math
+
+class BirthdayProblem(object):
+
+    def __init__(self):
+        pass
+
+
+def birthday_problem(problem_size=365):
+    birthdays = set()
+    while True:
+        new_birthday = random.randint(1, problem_size)
+        if new_birthday in birthdays:
+            return len(birthdays) + 1
+        birthdays.add(new_birthday)
+
+
+def theoretical_average(problem_size):
+    probabilities = []
+    contributions = []
+    for n in range(1, problem_size):
+        probability = (float(n-1) / problem_size) * falling_factorial_over_exponentiation(problem_size, n-1)
+        contribution = n * probability
+        probabilities.append(probability)
+        contributions.append(contribution)
+    return sum(contributions)
+
+
+def falling_factorial(n, k):
+    product = 1
+    while k > 0:
+        product *= n
+        n -= 1
+        k -= 1
+    return product
+
+
+
+def falling_factorial_over_exponentiation(n, k):
+    orig = n
+    product = float(1)
+    while k > 0:
+        product *= n
+        product = product/orig
+        n -= 1
+        k -= 1
+    return product
+
+
+def run_birthday_problem_n_times(times_to_run, problem_size=365):
+    return [birthday_problem(problem_size) for i in range(int(times_to_run))]
+
+
+def number_of_people_to_times_occured(runs):
+    number_of_people_to_times_occured = {}
+    for run in runs:
+        number_of_people_to_times_occured[run] = number_of_people_to_times_occured.get(run, 0) + 1
+
+
+if __name__ == '__main__':
+    times_to_run = 131072
+    while times_to_run <= 131072:
+        for problem_size in range(4000, 5000, 100):
+            average = sum(run_birthday_problem_n_times(times_to_run, problem_size=problem_size))/float(times_to_run)
+            print "problem size {3} ran {0} times, average was {1}, theoretical average is {2}".format(
+                times_to_run,
+                average,
+                theoretical_average(problem_size),
+                problem_size
+            )
+            print math.fabs(average - theoretical_average(problem_size))
+        times_to_run *= 2

dotfiles/lib/python/btree.py

@@ -0,0 +1,231 @@
+import bisect
+
+
+class BTreePrinter(object):
+
+    number_width = 4
+
+    subtree_space = object()
+
+    def __init__(self, btree):
+        self.btree = btree
+
+    def determine_width_of_node(self, node):
+        if node is None: return 0
+        return sum(map(self.determine_width_of_node, node.nodes)) + node.number_of_value_nodes * self.number_width
+
+    def determine_width_of_value_node(self, value_node):
+        return self.determine_width_of_node(value_node.node) if value_node.node is not None else self.number_width
+
+    def print_tree(self):
+        return self.print_levels_recursively([self.btree.head])
+
+    def print_levels_recursively(self, level):
+        if all(map(lambda x: x is self.subtree_space, level)): return
+        self.print_nodes_at_level(level)
+        print ''
+        self.print_levels_recursively(self.get_next_level(level))
+
+    def get_next_level(self, level):
+        new_level = []
+        for item in level:
+            if item is self.subtree_space:
+                new_level.append(item)
+            elif item is not None:
+                new_level.extend(item.nodes)
+                new_level.append(self.subtree_space)
+
+        return new_level
+
+    def print_nodes_at_level(self, level):
+        for item in level:
+            if item is self.subtree_space:
+                print ' ' * self.number_width,
+            else:
+                self.print_values_for_node(item)
+
+    def print_values_for_node(self, node):
+        if node is None: return
+        for value_node in node.value_nodes:
+            print ' ' * self.determine_width_of_node(value_node.node),
+            print '{num: ^{width}}'.format(num=value_node.value, width=self.number_width),
+        print (' ' * (self.determine_width_of_node(node.rightmost_node))),
+
+
+class IntegrityChecker(object):
+
+    def __init__(self, btree):
+        self.btree = btree
+
+    def check_integrity(self):
+        return self.check_for_items_smaller_in_right_subtree(self.btree.head) and self.check_for_unmatched_parents(self.btree.head)
+
+    def check_for_unmatched_parents(self, subtree):
+        if subtree is None:
+            return True
+
+        for node in subtree.nodes:
+            if node is None:
+                continue
+            if node.parent is not subtree:
+                return False
+            if not self.check_for_unmatched_parents(node):
+                return False
+        return True
+
+    def check_for_items_smaller_in_right_subtree(self, subtree):
+        if subtree is None:
+            return True
+
+        small_value = subtree.value_nodes[0].value
+        for value_node in subtree.value_nodes[1:]:
+            if not self.check_subtree_has_no_items_smaller_than(value_node.node, small_value):
+                return False
+
+            if not self.check_for_items_smaller_in_right_subtree(subtree.value_nodes[0].node):
+                return False
+
+        return self.check_subtree_has_no_items_smaller_than(subtree.rightmost_node, small_value)
+
+    def check_subtree_has_no_items_smaller_than(self, subtree, value):
+        if subtree is None:
+            return True
+        for value_node in subtree.value_nodes:
+            if value > value_node.value:
+                return False
+            if not self.check_subtree_has_no_items_smaller_than(value_node.node, value):
+                return False
+        return self.check_subtree_has_no_items_smaller_than(subtree.rightmost_node, value)
+
+
+class BTree(object):
+
+    @classmethod
+    def build_with_value(cls, value):
+        btree = cls()
+        btree.head = Node(btree, [ValueNode(value)])
+        return btree
+
+    def __init__(self):
+        self.head = None
+        self.inserted_items = []
+
+    def build_new_head(self, value_node):
+        new_rightmost_node = self.head
+        self.head = Node(self, [value_node])
+        value_node.node.parent = self.head
+        self.head.rightmost_node = new_rightmost_node
+        new_rightmost_node.parent = self.head
+        assert self.head.rightmost_node is not None
+        return value_node
+
+    def insert(self, value):
+        self.head.insert(value)
+        self.inserted_items.append(value)
+        self.head.check_integrity()
+        if not IntegrityChecker(self).check_integrity():
+            import ipdb; ipdb.set_trace()
+
+    promote_value_node = build_new_head
+
+    def __repr__(self):
+        return "BTree({0})".format(repr(self.head))
+
+
+class ValueNode(object):
+
+    def __init__(self, value, node=None):
+        self.value = value
+        self.node = node
+
+    def __lt__(self, other):
+        return self.value < other.value
+
+    def __gt__(self, other):
+        return self.value > other.value
+
+    def __repr__(self):
+        return "ValueNode({0}, {1})".format(repr(self.node), repr(self.value))
+
+
+class Node(object):
+
+    max_num_values = 3
+
+    def __init__(self, parent, value_nodes=None, rightmost_node=None):
+        self.parent = parent
+        self.value_nodes = value_nodes or []
+        self.rightmost_node = rightmost_node
+        self.claim_child_nodes()
+
+    def check_integrity(self):
+        if self.is_leaf_node: return True
+        if self.rightmost_node:
+            return all(child_node.check_integrity() for child_node in self.nodes if child_node is not None)
+        import ipdb; ipdb.set_trace()
+        return False
+
+    def claim_child_nodes(self):
+        for node in self.nodes:
+            if node:
+                node.parent = self
+
+    @property
+    def is_leaf_node(self):
+        return not any(self.nodes)
+
+    @property
+    def number_of_value_nodes(self):
+        return len(self.value_nodes)
+
+    @property
+    def nodes(self):
+        return [value_node.node for value_node in self.value_nodes] + [self.rightmost_node]
+
+    @property
+    def values(self):
+        return [value_node.value for value_node in self.value_nodes]
+
+    def __getitem__(self, item):
+        return self.nodes[item]
+
+    def promote_value_node(self, value_node):
+        bisect.insort(self.value_nodes, value_node)
+        if value_node.node:
+            value_node.node.parent = self
+        self.maybe_rebalance()
+
+    def maybe_rebalance(self):
+        if self.number_of_value_nodes < self.max_num_values:
+            return
+        value_node_to_promote = self.value_nodes[self.number_of_value_nodes/2]
+        promoted_nodes_old_node = value_node_to_promote.node
+        value_node_to_promote.node = Node(
+            self.parent,
+            value_nodes=self.value_nodes[:self.number_of_value_nodes/2],
+            rightmost_node=promoted_nodes_old_node
+        )
+        self.value_nodes = self.value_nodes[self.number_of_value_nodes/2+1:]
+        self.parent.promote_value_node(value_node_to_promote)
+        self.check_integrity()
+
+    def insert(self, value):
+        if self.is_leaf_node:
+            value_node = ValueNode(value)
+            bisect.insort(self.value_nodes, value_node)
+            self.maybe_rebalance()
+            return value_node
+
+        return self.pick_node(value).insert(value)
+
+    def pick_node(self, value):
+        if self.rightmost_node is None:
+            import ipdb; ipdb.set_trace()
+        for value_node in self.value_nodes:
+            if value < value_node.value:
+                return value_node.node
+
+        return self.rightmost_node
+
+    def __repr__(self):
+        return "Node({0}, {1})".format(", ".join(map(repr, self.value_nodes)), self.rightmost_node)

dotfiles/lib/python/but_one.py

@@ -0,0 +1,20 @@
+def left_partials(incoming):
+    product = 1
+    for i in incoming:
+        product *= i
+        yield product
+
+
+def but_one(incoming):
+    lpartials = list(left_partials(incoming))
+    rproduct = 1
+    result = [None]*len(incoming)
+    for i in range(len(incoming)):
+        back_index = len(incoming) - i - 1
+        if back_index > 0:
+            result[back_index] = rproduct * lpartials[back_index-1]
+            if back_index < len(incoming):
+                rproduct *= incoming[back_index]
+        else:
+            result[back_index] = rproduct
+    return result

dotfiles/lib/python/children_problem.py

@@ -0,0 +1,20 @@
+import operator
+
+def generate_decreasing_n_sequence_with_bounded_sum(sequence_length, sum_bound, value_bound=float('inf')):
+    if sequence_length == 0:
+        yield []
+        return
+    min_remaining = sequence_length*(sequence_length - 1)/2
+    bound_for_current = min(sum_bound - min_remaining, value_bound)
+
+    for value in range(sequence_length, bound_for_current):
+        for sequence in generate_decreasing_n_sequence_with_bounded_sum(sequence_length - 1, sum_bound - value, value_bound=value):
+            yield [value] + sequence
+
+
+def build_products_to_sequences_map():
+    product_to_sequences_map = {}
+    for sequence in generate_decreasing_n_sequence_with_bounded_sum(4, 18):
+        product = reduce(operator.mul, sequence, 1)
+        product_to_sequences_map.setdefault(product, []).append(sequence)
+    return product_to_sequences_map

dotfiles/lib/python/tasks.py

@@ -0,0 +1,27 @@
+import datetime
+
+from invoke import task, run
+
+
+@task
+def histogram(ignore=''):
+    result = run('git rev-list --all')
+    date_to_adds = {}
+    date_to_deletes = {}
+    for sha in result.stdout.split('\n'):
+        result = run('git diff-excluding {1} {0}~1 {0} --numstat'.format(sha, ignore), hide=True)
+        added, deleted = get_total(result.stdout)
+        iso8601 = run('git log {0} --pretty=format:%ai -1'.format(sha), hide=True).stdout.strip()
+        commit_date = datetime.datetime.strptime(iso8601, "%Y-%m-%dT%H:%M:%S %z").date()
+        date_to_adds[commit_date] = date_to_adds.get(commit_date) + added
+        date_to_deletes[commit_date] = date_to_deletes.get(commit_date) + deleted
+    print date_to_adds
+    print date_to_deletes
+
+
+def get_total(output):
+    try:
+        return sum(int(line.split()[0]) for line in output.split('\n')), sum(int(line.split()[1]) for line in output.split('\n'))
+    except:
+        import ipdb; ipdb.set_trace()
+

dotfiles/lib/python/tox2travis.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python
+from tox._config import parseconfig
+from yaml import dump
+
+
+class TravisFromTox(object):
+
+    def __init__(self, tox_config):
+        self._tox_config = tox_config
+
+    def build_travis_dict(self):
+        return {
+            'language': 'python',
+            'install': ['pip install "tox>=1.8.0"'],
+            'script': 'tox',
+            'env': self._get_environment_variables()
+        }
+
+    def _get_environment_variables(self):
+        return ['TOXENV={0}'.format(env) for env in self._tox_config.envlist]
+
+    def build_travis_yaml(self):
+        return dump(self.build_travis_dict(), default_flow_style=False)
+
+
+if __name__ == '__main__':
+    print TravisFromTox(parseconfig()).build_travis_yaml()

dotfiles/lib/zsh/history.sh

@@ -3,12 +3,12 @@ HISTFILE="$(readlink -f $HOME/.zsh_history)"
 HISTSIZE=100000
 SAVEHIST=100000
 
+setopt share_history # share command history data
 setopt extended_history
 setopt hist_expire_dups_first
 setopt hist_ignore_dups # ignore duplication command history list
 setopt hist_ignore_space
 setopt hist_verify
-setopt share_history # share command history data
 
 # TODO: Explain what some of this does..