prometheus-client-mmap 0.9.8 → 0.12.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 699286c1c257884747f07061bf5dcb7c078720b5c8c81826965e826b3c3e727a
-  data.tar.gz: 61d2caba4d09883699471a4c631a60168fed152d38a59ff069be08527a3f65f5
+  metadata.gz: 3fec107115b6e90ebfbc109f0603bb44ca7d7165e637a2e6448ce463ab54c3c7
+  data.tar.gz: 9e9e0382069f57aebe0fb35508a1b82a8e1c61b4b228be573807241b2a434fb8
 SHA512:
-  metadata.gz: ac048db43f14bc544fbd199b178e62536b6a4f2ebb1e13f33d7540d847bf3c0e44ee9af97693a26ea3b227aebc29568a9fbedce8fac467e591329090ce6aadbf
-  data.tar.gz: 5185dbbd600e6fb91615361a5621feefb1ac2fd65f90439a9daf9e7b69d59a72e083ef7d174b97f92109056396c5afccfd0961a9c6d6bcc52267e43f9db9c19b
+  metadata.gz: 6747ece05958e5b65a92e7305e8edb167e7ffafa84d5f0b24ccc56e8c8977add781f7c88eab1ecca117ae88bb0bbd7150273307f0c5314f33249246cef317dbf
+  data.tar.gz: 7454daa53c15a7cf57e359ebec87dda164609ad43620ebf8a72e287cac4113caba08de8ca3ba2d1efdcf1c0a05f2bdbfb7c2b524d8eb27e63444844f06469ed5

data/ext/fast_mmaped_file/fast_mmaped_file.c

@@ -1,20 +1,17 @@
 #include <errno.h>
+#include <hashmap.h>
+#include <jsmn.h>
 #include <ruby.h>
 #include <ruby/intern.h>
-
 #include <sys/mman.h>
 
-#include <hashmap.h>
-#include <jsmn.h>
-
-#include "globals.h"
-#include "utils.h"
-#include "value_access.h"
-
 #include "file_parsing.h"
 #include "file_reading.h"
+#include "globals.h"
 #include "mmap.h"
 #include "rendering.h"
+#include "utils.h"
+#include "value_access.h"
 
 VALUE MMAPED_FILE = Qnil;
 

data/ext/fast_mmaped_file/file_parsing.c

@@ -1,9 +1,9 @@
+#include "file_parsing.h"
+
 #include <hashmap.h>
 #include <jsmn.h>
-#include <ruby.h>
 
 #include "file_format.h"
-#include "file_parsing.h"
 #include "globals.h"
 #include "utils.h"
 

data/ext/fast_mmaped_file/file_parsing.h

@@ -2,6 +2,7 @@
 #define FILE_PARSING_H
 #include <file_reading.h>
 #include <hashmap.h>
+#include <ruby.h>
 
 typedef struct {
     char *json;

data/ext/fast_mmaped_file/file_reading.c

@@ -1,8 +1,8 @@
+#include "file_reading.h"
+
 #include <errno.h>
 #include <fcntl.h>
-#include <ruby.h>
 
-#include "file_reading.h"
 #include "utils.h"
 
 static int file_open(file_t *source, const char *filepath) {

data/ext/fast_mmaped_file/hashmap.c

@@ -0,0 +1,692 @@
+/*
+ * Copyright (c) 2016-2017 David Leeds <davidesleeds@gmail.com>
+ *
+ * Hashmap is free software; you can redistribute it and/or modify
+ * it under the terms of the MIT license. See LICENSE for details.
+ */
+
+#include <stdlib.h>
+#include <stdint.h>
+#include <stdbool.h>
+#include <string.h>
+#include <errno.h>
+
+#include "hashmap.h"
+
+#ifndef HASHMAP_NOASSERT
+#include <assert.h>
+#define HASHMAP_ASSERT(expr)	assert(expr)
+#else
+#define HASHMAP_ASSERT(expr)
+#endif
+
+/* Table sizes must be powers of 2 */
+#define HASHMAP_SIZE_MIN		(1 << 5)	/* 32 */
+#define HASHMAP_SIZE_DEFAULT		(1 << 8)	/* 256 */
+#define HASHMAP_SIZE_MOD(map, val)	((val) & ((map)->table_size - 1))
+
+/* Limit for probing is 1/2 of table_size */
+#define HASHMAP_PROBE_LEN(map)		((map)->table_size >> 1)
+/* Return the next linear probe index */
+#define HASHMAP_PROBE_NEXT(map, index)	 HASHMAP_SIZE_MOD(map, (index) + 1)
+
+/* Check if index b is less than or equal to index a */
+#define HASHMAP_INDEX_LE(map, a, b)	\
+	((a) == (b) || (((b) - (a)) & ((map)->table_size >> 1)) != 0)
+
+
+struct hashmap_entry {
+	void *key;
+	void *data;
+#ifdef HASHMAP_METRICS
+	size_t num_collisions;
+#endif
+};
+
+
+/*
+ * Enforce a maximum 0.75 load factor.
+ */
+static inline size_t hashmap_table_min_size_calc(size_t num_entries)
+{
+	return num_entries + (num_entries / 3);
+}
+
+/*
+ * Calculate the optimal table size, given the specified max number
+ * of elements.
+ */
+static size_t hashmap_table_size_calc(size_t num_entries)
+{
+	size_t table_size;
+	size_t min_size;
+
+	table_size = hashmap_table_min_size_calc(num_entries);
+
+	/* Table size is always a power of 2 */
+	min_size = HASHMAP_SIZE_MIN;
+	while (min_size < table_size) {
+		min_size <<= 1;
+	}
+	return min_size;
+}
+
+/*
+ * Get a valid hash table index from a key.
+ */
+static inline size_t hashmap_calc_index(const struct hashmap *map,
+	const void *key)
+{
+	return HASHMAP_SIZE_MOD(map, map->hash(key));
+}
+
+/*
+ * Return the next populated entry, starting with the specified one.
+ * Returns NULL if there are no more valid entries.
+ */
+static struct hashmap_entry *hashmap_entry_get_populated(
+	const struct hashmap *map, struct hashmap_entry *entry)
+{
+	for (; entry < &map->table[map->table_size]; ++entry) {
+		if (entry->key) {
+			return entry;
+		}
+	}
+	return NULL;
+}
+
+/*
+ * Find the hashmap entry with the specified key, or an empty slot.
+ * Returns NULL if the entire table has been searched without finding a match.
+ */
+static struct hashmap_entry *hashmap_entry_find(const struct hashmap *map,
+	const void *key, bool find_empty)
+{
+	size_t i;
+	size_t index;
+	size_t probe_len = HASHMAP_PROBE_LEN(map);
+	struct hashmap_entry *entry;
+
+	index = hashmap_calc_index(map, key);
+
+	/* Linear probing */
+	for (i = 0; i < probe_len; ++i) {
+		entry = &map->table[index];
+		if (!entry->key) {
+			if (find_empty) {
+#ifdef HASHMAP_METRICS
+				entry->num_collisions = i;
+#endif
+				return entry;
+			}
+			return NULL;
+		}
+		if (map->key_compare(key, entry->key) == 0) {
+			return entry;
+		}
+		index = HASHMAP_PROBE_NEXT(map, index);
+	}
+	return NULL;
+}
+
+/*
+ * Removes the specified entry and processes the proceeding entries to reduce
+ * the load factor and keep the chain continuous.  This is a required
+ * step for hash maps using linear probing.
+ */
+static void hashmap_entry_remove(struct hashmap *map,
+	struct hashmap_entry *removed_entry)
+{
+	size_t i;
+#ifdef HASHMAP_METRICS
+	size_t removed_i = 0;
+#endif
+	size_t index;
+	size_t entry_index;
+	size_t removed_index = (removed_entry - map->table);
+	struct hashmap_entry *entry;
+
+	/* Free the key */
+	if (map->key_free) {
+		map->key_free(removed_entry->key);
+	}
+	--map->num_entries;
+
+	/* Fill the free slot in the chain */
+	index = HASHMAP_PROBE_NEXT(map, removed_index);
+	for (i = 1; i < map->table_size; ++i) {
+		entry = &map->table[index];
+		if (!entry->key) {
+			/* Reached end of chain */
+			break;
+		}
+		entry_index = hashmap_calc_index(map, entry->key);
+		/* Shift in entries with an index <= to the removed slot */
+		if (HASHMAP_INDEX_LE(map, removed_index, entry_index)) {
+#ifdef HASHMAP_METRICS
+			entry->num_collisions -= (i - removed_i);
+			removed_i = i;
+#endif
+			memcpy(removed_entry, entry, sizeof(*removed_entry));
+			removed_index = index;
+			removed_entry = entry;
+		}
+		index = HASHMAP_PROBE_NEXT(map, index);
+	}
+	/* Clear the last removed entry */
+	memset(removed_entry, 0, sizeof(*removed_entry));
+}
+
+/*
+ * Reallocates the hash table to the new size and rehashes all entries.
+ * new_size MUST be a power of 2.
+ * Returns 0 on success and -errno on allocation or hash function failure.
+ */
+static int hashmap_rehash(struct hashmap *map, size_t new_size)
+{
+	size_t old_size;
+	struct hashmap_entry *old_table;
+	struct hashmap_entry *new_table;
+	struct hashmap_entry *entry;
+	struct hashmap_entry *new_entry;
+
+	HASHMAP_ASSERT(new_size >= HASHMAP_SIZE_MIN);
+	HASHMAP_ASSERT((new_size & (new_size - 1)) == 0);
+
+	new_table = (struct hashmap_entry *)calloc(new_size,
+	    sizeof(struct hashmap_entry));
+	if (!new_table) {
+		return -ENOMEM;
+	}
+	/* Backup old elements in case of rehash failure */
+	old_size = map->table_size;
+	old_table = map->table;
+	map->table_size = new_size;
+	map->table = new_table;
+	/* Rehash */
+	for (entry = old_table; entry < &old_table[old_size]; ++entry) {
+		if (!entry->data) {
+			/* Only copy entries with data */
+			continue;
+		}
+		new_entry = hashmap_entry_find(map, entry->key, true);
+		if (!new_entry) {
+			/*
+			 * The load factor is too high with the new table
+			 * size, or a poor hash function was used.
+			 */
+			goto revert;
+		}
+		/* Shallow copy (intentionally omits num_collisions) */
+		new_entry->key = entry->key;
+		new_entry->data = entry->data;
+	}
+	free(old_table);
+	return 0;
+revert:
+	map->table_size = old_size;
+	map->table = old_table;
+	free(new_table);
+	return -EINVAL;
+}
+
+/*
+ * Iterate through all entries and free all keys.
+ */
+static void hashmap_free_keys(struct hashmap *map)
+{
+	struct hashmap_iter *iter;
+
+	if (!map->key_free) {
+		return;
+	}
+	for (iter = hashmap_iter(map); iter;
+	    iter = hashmap_iter_next(map, iter)) {
+		map->key_free((void *)hashmap_iter_get_key(iter));
+	}
+}
+
+/*
+ * Initialize an empty hashmap.  A hash function and a key comparator are
+ * required.
+ *
+ * hash_func should return an even distribution of numbers between 0
+ * and SIZE_MAX varying on the key provided.
+ *
+ * key_compare_func should return 0 if the keys match, and non-zero otherwise.
+ *
+ * initial_size is optional, and may be set to the max number of entries
+ * expected to be put in the hash table.  This is used as a hint to
+ * pre-allocate the hash table to the minimum size needed to avoid
+ * gratuitous rehashes.  If initial_size 0, a default size will be used.
+ *
+ * Returns 0 on success and -errno on failure.
+ */
+int hashmap_init(struct hashmap *map, size_t (*hash_func)(const void *),
+	int (*key_compare_func)(const void *, const void *),
+	size_t initial_size)
+{
+	HASHMAP_ASSERT(map != NULL);
+	HASHMAP_ASSERT(hash_func != NULL);
+	HASHMAP_ASSERT(key_compare_func != NULL);
+
+	if (!initial_size) {
+		initial_size = HASHMAP_SIZE_DEFAULT;
+	} else {
+		/* Convert init size to valid table size */
+		initial_size = hashmap_table_size_calc(initial_size);
+	}
+	map->table_size_init = initial_size;
+	map->table_size = initial_size;
+	map->num_entries = 0;
+	map->table = (struct hashmap_entry *)calloc(initial_size,
+	    sizeof(struct hashmap_entry));
+	if (!map->table) {
+		return -ENOMEM;
+	}
+	map->hash = hash_func;
+	map->key_compare = key_compare_func;
+	map->key_alloc = NULL;
+	map->key_free = NULL;
+	return 0;
+}
+
+/*
+ * Free the hashmap and all associated memory.
+ */
+void hashmap_destroy(struct hashmap *map)
+{
+	if (!map) {
+		return;
+	}
+	hashmap_free_keys(map);
+	free(map->table);
+	memset(map, 0, sizeof(*map));
+}
+
+/*
+ * Enable internal memory management of hash keys.
+ */
+void hashmap_set_key_alloc_funcs(struct hashmap *map,
+	void *(*key_alloc_func)(const void *),
+	void (*key_free_func)(void *))
+{
+	HASHMAP_ASSERT(map != NULL);
+
+	map->key_alloc = key_alloc_func;
+	map->key_free = key_free_func;
+}
+
+/*
+ * Add an entry to the hashmap.  If an entry with a matching key already
+ * exists and has a data pointer associated with it, the existing data
+ * pointer is returned, instead of assigning the new value.  Compare
+ * the return value with the data passed in to determine if a new entry was
+ * created.  Returns NULL if memory allocation failed.
+ */
+void *hashmap_put(struct hashmap *map, const void *key, void *data)
+{
+	struct hashmap_entry *entry;
+
+	HASHMAP_ASSERT(map != NULL);
+	HASHMAP_ASSERT(key != NULL);
+
+	/* Rehash with 2x capacity if load factor is approaching 0.75 */
+	if (map->table_size <= hashmap_table_min_size_calc(map->num_entries)) {
+		hashmap_rehash(map, map->table_size << 1);
+	}
+	entry = hashmap_entry_find(map, key, true);
+	if (!entry) {
+		/*
+		 * Cannot find an empty slot.  Either out of memory, or using
+		 * a poor hash function.  Attempt to rehash once to reduce
+		 * chain length.
+		 */
+		if (hashmap_rehash(map, map->table_size << 1) < 0) {
+			return NULL;
+		}
+		entry = hashmap_entry_find(map, key, true);
+		if (!entry) {
+			return NULL;
+		}
+	}
+	if (!entry->key) {
+		/* Allocate copy of key to simplify memory management */
+		if (map->key_alloc) {
+			entry->key = map->key_alloc(key);
+			if (!entry->key) {
+				return NULL;
+			}
+		} else {
+			entry->key = (void *)key;
+		}
+		++map->num_entries;
+	} else if (entry->data) {
+		/* Do not overwrite existing data */
+		return entry->data;
+	}
+	entry->data = data;
+	return data;
+}
+
+/*
+ * Return the data pointer, or NULL if no entry exists.
+ */
+void *hashmap_get(const struct hashmap *map, const void *key)
+{
+	struct hashmap_entry *entry;
+
+	HASHMAP_ASSERT(map != NULL);
+	HASHMAP_ASSERT(key != NULL);
+
+	entry = hashmap_entry_find(map, key, false);
+	if (!entry) {
+		return NULL;
+	}
+	return entry->data;
+}
+
+/*
+ * Remove an entry with the specified key from the map.
+ * Returns the data pointer, or NULL, if no entry was found.
+ */
+void *hashmap_remove(struct hashmap *map, const void *key)
+{
+	struct hashmap_entry *entry;
+	void *data;
+
+	HASHMAP_ASSERT(map != NULL);
+	HASHMAP_ASSERT(key != NULL);
+
+	entry = hashmap_entry_find(map, key, false);
+	if (!entry) {
+		return NULL;
+	}
+	data = entry->data;
+	/* Clear the entry and make the chain contiguous */
+	hashmap_entry_remove(map, entry);
+	return data;
+}
+
+/*
+ * Remove all entries.
+ */
+void hashmap_clear(struct hashmap *map)
+{
+	HASHMAP_ASSERT(map != NULL);
+
+	hashmap_free_keys(map);
+	map->num_entries = 0;
+	memset(map->table, 0, sizeof(struct hashmap_entry) * map->table_size);
+}
+
+/*
+ * Remove all entries and reset the hash table to its initial size.
+ */
+void hashmap_reset(struct hashmap *map)
+{
+	struct hashmap_entry *new_table;
+
+	HASHMAP_ASSERT(map != NULL);
+
+	hashmap_clear(map);
+	if (map->table_size == map->table_size_init) {
+		return;
+	}
+	new_table = (struct hashmap_entry *)realloc(map->table,
+	    sizeof(struct hashmap_entry) * map->table_size_init);
+	if (!new_table) {
+		return;
+	}
+	map->table = new_table;
+	map->table_size = map->table_size_init;
+}
+
+/*
+ * Return the number of entries in the hash map.
+ */
+size_t hashmap_size(const struct hashmap *map)
+{
+	HASHMAP_ASSERT(map != NULL);
+
+	return map->num_entries;
+}
+
+/*
+ * Get a new hashmap iterator.  The iterator is an opaque
+ * pointer that may be used with hashmap_iter_*() functions.
+ * Hashmap iterators are INVALID after a put or remove operation is performed.
+ * hashmap_iter_remove() allows safe removal during iteration.
+ */
+struct hashmap_iter *hashmap_iter(const struct hashmap *map)
+{
+	HASHMAP_ASSERT(map != NULL);
+
+	if (!map->num_entries) {
+		return NULL;
+	}
+	return (struct hashmap_iter *)hashmap_entry_get_populated(map,
+		map->table);
+}
+
+/*
+ * Return an iterator to the next hashmap entry.  Returns NULL if there are
+ * no more entries.
+ */
+struct hashmap_iter *hashmap_iter_next(const struct hashmap *map,
+	const struct hashmap_iter *iter)
+{
+	struct hashmap_entry *entry = (struct hashmap_entry *)iter;
+
+	HASHMAP_ASSERT(map != NULL);
+
+	if (!iter) {
+		return NULL;
+	}
+	return (struct hashmap_iter *)hashmap_entry_get_populated(map,
+		entry + 1);
+}
+
+/*
+ * Remove the hashmap entry pointed to by this iterator and return an
+ * iterator to the next entry.  Returns NULL if there are no more entries.
+ */
+struct hashmap_iter *hashmap_iter_remove(struct hashmap *map,
+	const struct hashmap_iter *iter)
+{
+	struct hashmap_entry *entry = (struct hashmap_entry *)iter;
+
+	HASHMAP_ASSERT(map != NULL);
+
+	if (!iter) {
+		return NULL;
+	}
+	if (!entry->key) {
+		/* Iterator is invalid, so just return the next valid entry */
+		return hashmap_iter_next(map, iter);
+	}
+	hashmap_entry_remove(map, entry);
+	return (struct hashmap_iter *)hashmap_entry_get_populated(map, entry);
+}
+
+/*
+ * Return the key of the entry pointed to by the iterator.
+ */
+const void *hashmap_iter_get_key(const struct hashmap_iter *iter)
+{
+	if (!iter) {
+		return NULL;
+	}
+	return (const void *)((struct hashmap_entry *)iter)->key;
+}
+
+/*
+ * Return the data of the entry pointed to by the iterator.
+ */
+void *hashmap_iter_get_data(const struct hashmap_iter *iter)
+{
+	if (!iter) {
+		return NULL;
+	}
+	return ((struct hashmap_entry *)iter)->data;
+}
+
+/*
+ * Set the data pointer of the entry pointed to by the iterator.
+ */
+void hashmap_iter_set_data(const struct hashmap_iter *iter, void *data)
+{
+	if (!iter) {
+		return;
+	}
+	((struct hashmap_entry *)iter)->data = data;
+}
+
+/*
+ * Invoke func for each entry in the hashmap.  Unlike the hashmap_iter_*()
+ * interface, this function supports calls to hashmap_remove() during iteration.
+ * However, it is an error to put or remove an entry other than the current one,
+ * and doing so will immediately halt iteration and return an error.
+ * Iteration is stopped if func returns non-zero.  Returns func's return
+ * value if it is < 0, otherwise, 0.
+ */
+int hashmap_foreach(const struct hashmap *map,
+	int (*func)(const void *, void *, void *), void *arg)
+{
+	struct hashmap_entry *entry;
+	size_t num_entries;
+	const void *key;
+	int rc;
+
+	HASHMAP_ASSERT(map != NULL);
+	HASHMAP_ASSERT(func != NULL);
+
+	entry = map->table;
+	for (entry = map->table; entry < &map->table[map->table_size];
+	    ++entry) {
+		if (!entry->key) {
+			continue;
+		}
+		num_entries = map->num_entries;
+		key = entry->key;
+		rc = func(entry->key, entry->data, arg);
+		if (rc < 0) {
+			return rc;
+		}
+		if (rc > 0) {
+			return 0;
+		}
+		/* Run this entry again if func() deleted it */
+		if (entry->key != key) {
+			--entry;
+		} else if (num_entries != map->num_entries) {
+			/* Stop immediately if func put/removed another entry */
+			return -1;
+		}
+	}
+	return 0;
+}
+
+/*
+ * Default hash function for string keys.
+ * This is an implementation of the well-documented Jenkins one-at-a-time
+ * hash function.
+ */
+size_t hashmap_hash_string(const void *key)
+{
+	const char *key_str = (const char *)key;
+	size_t hash = 0;
+
+	for (; *key_str; ++key_str) {
+		hash += *key_str;
+		hash += (hash << 10);
+		hash ^= (hash >> 6);
+	}
+	hash += (hash << 3);
+	hash ^= (hash >> 11);
+	hash += (hash << 15);
+	return hash;
+}
+
+/*
+ * Default key comparator function for string keys.
+ */
+int hashmap_compare_string(const void *a, const void *b)
+{
+	return strcmp((const char *)a, (const char *)b);
+}
+
+/*
+ * Default key allocation function for string keys.  Use free() for the
+ * key_free_func.
+ */
+void *hashmap_alloc_key_string(const void *key)
+{
+	return (void *)strdup((const char *)key);
+}
+
+#ifdef HASHMAP_METRICS
+/*
+ * Return the load factor.
+ */
+double hashmap_load_factor(const struct hashmap *map)
+{
+	HASHMAP_ASSERT(map != NULL);
+
+	if (!map->table_size) {
+		return 0;
+	}
+	return (double)map->num_entries / map->table_size;
+}
+
+/*
+ * Return the average number of collisions per entry.
+ */
+double hashmap_collisions_mean(const struct hashmap *map)
+{
+	struct hashmap_entry *entry;
+	size_t total_collisions = 0;
+
+	HASHMAP_ASSERT(map != NULL);
+
+	if (!map->num_entries) {
+		return 0;
+	}
+	for (entry = map->table; entry < &map->table[map->table_size];
+	    ++entry) {
+		if (!entry->key) {
+			continue;
+		}
+		total_collisions += entry->num_collisions;
+	}
+	return (double)total_collisions / map->num_entries;
+}
+
+/*
+ * Return the variance between entry collisions.  The higher the variance,
+ * the more likely the hash function is poor and is resulting in clustering.
+ */
+double hashmap_collisions_variance(const struct hashmap *map)
+{
+	struct hashmap_entry *entry;
+	double mean_collisions;
+	double variance;
+	double total_variance = 0;
+
+	HASHMAP_ASSERT(map != NULL);
+
+	if (!map->num_entries) {
+		return 0;
+	}
+	mean_collisions = hashmap_collisions_mean(map);
+	for (entry = map->table; entry < &map->table[map->table_size];
+	    ++entry) {
+		if (!entry->key) {
+			continue;
+		}
+		variance = (double)entry->num_collisions - mean_collisions;
+		total_variance += variance * variance;
+	}
+	return total_variance / map->num_entries;
+}
+#endif

data/ext/fast_mmaped_file/jsmn.c

@@ -0,0 +1,314 @@
+#include "jsmn.h"
+
+/**
+ * Allocates a fresh unused token from the token pull.
+ */
+static jsmntok_t *jsmn_alloc_token(jsmn_parser *parser,
+		jsmntok_t *tokens, size_t num_tokens) {
+	jsmntok_t *tok;
+	if (parser->toknext >= num_tokens) {
+		return NULL;
+	}
+	tok = &tokens[parser->toknext++];
+	tok->start = tok->end = -1;
+	tok->size = 0;
+#ifdef JSMN_PARENT_LINKS
+	tok->parent = -1;
+#endif
+	return tok;
+}
+
+/**
+ * Fills token type and boundaries.
+ */
+static void jsmn_fill_token(jsmntok_t *token, jsmntype_t type,
+                            int start, int end) {
+	token->type = type;
+	token->start = start;
+	token->end = end;
+	token->size = 0;
+}
+
+/**
+ * Fills next available token with JSON primitive.
+ */
+static int jsmn_parse_primitive(jsmn_parser *parser, const char *js,
+		size_t len, jsmntok_t *tokens, size_t num_tokens) {
+	jsmntok_t *token;
+	int start;
+
+	start = parser->pos;
+
+	for (; parser->pos < len && js[parser->pos] != '\0'; parser->pos++) {
+		switch (js[parser->pos]) {
+#ifndef JSMN_STRICT
+			/* In strict mode primitive must be followed by "," or "}" or "]" */
+			case ':':
+#endif
+			case '\t' : case '\r' : case '\n' : case ' ' :
+			case ','  : case ']'  : case '}' :
+				goto found;
+		}
+		if (js[parser->pos] < 32 || js[parser->pos] >= 127) {
+			parser->pos = start;
+			return JSMN_ERROR_INVAL;
+		}
+	}
+#ifdef JSMN_STRICT
+	/* In strict mode primitive must be followed by a comma/object/array */
+	parser->pos = start;
+	return JSMN_ERROR_PART;
+#endif
+
+found:
+	if (tokens == NULL) {
+		parser->pos--;
+		return 0;
+	}
+	token = jsmn_alloc_token(parser, tokens, num_tokens);
+	if (token == NULL) {
+		parser->pos = start;
+		return JSMN_ERROR_NOMEM;
+	}
+	jsmn_fill_token(token, JSMN_PRIMITIVE, start, parser->pos);
+#ifdef JSMN_PARENT_LINKS
+	token->parent = parser->toksuper;
+#endif
+	parser->pos--;
+	return 0;
+}
+
+/**
+ * Fills next token with JSON string.
+ */
+static int jsmn_parse_string(jsmn_parser *parser, const char *js,
+		size_t len, jsmntok_t *tokens, size_t num_tokens) {
+	jsmntok_t *token;
+
+	int start = parser->pos;
+
+	parser->pos++;
+
+	/* Skip starting quote */
+	for (; parser->pos < len && js[parser->pos] != '\0'; parser->pos++) {
+		char c = js[parser->pos];
+
+		/* Quote: end of string */
+		if (c == '\"') {
+			if (tokens == NULL) {
+				return 0;
+			}
+			token = jsmn_alloc_token(parser, tokens, num_tokens);
+			if (token == NULL) {
+				parser->pos = start;
+				return JSMN_ERROR_NOMEM;
+			}
+			jsmn_fill_token(token, JSMN_STRING, start+1, parser->pos);
+#ifdef JSMN_PARENT_LINKS
+			token->parent = parser->toksuper;
+#endif
+			return 0;
+		}
+
+		/* Backslash: Quoted symbol expected */
+		if (c == '\\' && parser->pos + 1 < len) {
+			int i;
+			parser->pos++;
+			switch (js[parser->pos]) {
+				/* Allowed escaped symbols */
+				case '\"': case '/' : case '\\' : case 'b' :
+				case 'f' : case 'r' : case 'n'  : case 't' :
+					break;
+				/* Allows escaped symbol \uXXXX */
+				case 'u':
+					parser->pos++;
+					for(i = 0; i < 4 && parser->pos < len && js[parser->pos] != '\0'; i++) {
+						/* If it isn't a hex character we have an error */
+						if(!((js[parser->pos] >= 48 && js[parser->pos] <= 57) || /* 0-9 */
+									(js[parser->pos] >= 65 && js[parser->pos] <= 70) || /* A-F */
+									(js[parser->pos] >= 97 && js[parser->pos] <= 102))) { /* a-f */
+							parser->pos = start;
+							return JSMN_ERROR_INVAL;
+						}
+						parser->pos++;
+					}
+					parser->pos--;
+					break;
+				/* Unexpected symbol */
+				default:
+					parser->pos = start;
+					return JSMN_ERROR_INVAL;
+			}
+		}
+	}
+	parser->pos = start;
+	return JSMN_ERROR_PART;
+}
+
+/**
+ * Parse JSON string and fill tokens.
+ */
+int jsmn_parse(jsmn_parser *parser, const char *js, size_t len,
+		jsmntok_t *tokens, unsigned int num_tokens) {
+	int r;
+	int i;
+	jsmntok_t *token;
+	int count = parser->toknext;
+
+	for (; parser->pos < len && js[parser->pos] != '\0'; parser->pos++) {
+		char c;
+		jsmntype_t type;
+
+		c = js[parser->pos];
+		switch (c) {
+			case '{': case '[':
+				count++;
+				if (tokens == NULL) {
+					break;
+				}
+				token = jsmn_alloc_token(parser, tokens, num_tokens);
+				if (token == NULL)
+					return JSMN_ERROR_NOMEM;
+				if (parser->toksuper != -1) {
+					tokens[parser->toksuper].size++;
+#ifdef JSMN_PARENT_LINKS
+					token->parent = parser->toksuper;
+#endif
+				}
+				token->type = (c == '{' ? JSMN_OBJECT : JSMN_ARRAY);
+				token->start = parser->pos;
+				parser->toksuper = parser->toknext - 1;
+				break;
+			case '}': case ']':
+				if (tokens == NULL)
+					break;
+				type = (c == '}' ? JSMN_OBJECT : JSMN_ARRAY);
+#ifdef JSMN_PARENT_LINKS
+				if (parser->toknext < 1) {
+					return JSMN_ERROR_INVAL;
+				}
+				token = &tokens[parser->toknext - 1];
+				for (;;) {
+					if (token->start != -1 && token->end == -1) {
+						if (token->type != type) {
+							return JSMN_ERROR_INVAL;
+						}
+						token->end = parser->pos + 1;
+						parser->toksuper = token->parent;
+						break;
+					}
+					if (token->parent == -1) {
+						if(token->type != type || parser->toksuper == -1) {
+							return JSMN_ERROR_INVAL;
+						}
+						break;
+					}
+					token = &tokens[token->parent];
+				}
+#else
+				for (i = parser->toknext - 1; i >= 0; i--) {
+					token = &tokens[i];
+					if (token->start != -1 && token->end == -1) {
+						if (token->type != type) {
+							return JSMN_ERROR_INVAL;
+						}
+						parser->toksuper = -1;
+						token->end = parser->pos + 1;
+						break;
+					}
+				}
+				/* Error if unmatched closing bracket */
+				if (i == -1) return JSMN_ERROR_INVAL;
+				for (; i >= 0; i--) {
+					token = &tokens[i];
+					if (token->start != -1 && token->end == -1) {
+						parser->toksuper = i;
+						break;
+					}
+				}
+#endif
+				break;
+			case '\"':
+				r = jsmn_parse_string(parser, js, len, tokens, num_tokens);
+				if (r < 0) return r;
+				count++;
+				if (parser->toksuper != -1 && tokens != NULL)
+					tokens[parser->toksuper].size++;
+				break;
+			case '\t' : case '\r' : case '\n' : case ' ':
+				break;
+			case ':':
+				parser->toksuper = parser->toknext - 1;
+				break;
+			case ',':
+				if (tokens != NULL && parser->toksuper != -1 &&
+						tokens[parser->toksuper].type != JSMN_ARRAY &&
+						tokens[parser->toksuper].type != JSMN_OBJECT) {
+#ifdef JSMN_PARENT_LINKS
+					parser->toksuper = tokens[parser->toksuper].parent;
+#else
+					for (i = parser->toknext - 1; i >= 0; i--) {
+						if (tokens[i].type == JSMN_ARRAY || tokens[i].type == JSMN_OBJECT) {
+							if (tokens[i].start != -1 && tokens[i].end == -1) {
+								parser->toksuper = i;
+								break;
+							}
+						}
+					}
+#endif
+				}
+				break;
+#ifdef JSMN_STRICT
+			/* In strict mode primitives are: numbers and booleans */
+			case '-': case '0': case '1' : case '2': case '3' : case '4':
+			case '5': case '6': case '7' : case '8': case '9':
+			case 't': case 'f': case 'n' :
+				/* And they must not be keys of the object */
+				if (tokens != NULL && parser->toksuper != -1) {
+					jsmntok_t *t = &tokens[parser->toksuper];
+					if (t->type == JSMN_OBJECT ||
+							(t->type == JSMN_STRING && t->size != 0)) {
+						return JSMN_ERROR_INVAL;
+					}
+				}
+#else
+			/* In non-strict mode every unquoted value is a primitive */
+			default:
+#endif
+				r = jsmn_parse_primitive(parser, js, len, tokens, num_tokens);
+				if (r < 0) return r;
+				count++;
+				if (parser->toksuper != -1 && tokens != NULL)
+					tokens[parser->toksuper].size++;
+				break;
+
+#ifdef JSMN_STRICT
+			/* Unexpected char in strict mode */
+			default:
+				return JSMN_ERROR_INVAL;
+#endif
+		}
+	}
+
+	if (tokens != NULL) {
+		for (i = parser->toknext - 1; i >= 0; i--) {
+			/* Unmatched opened object or array */
+			if (tokens[i].start != -1 && tokens[i].end == -1) {
+				return JSMN_ERROR_PART;
+			}
+		}
+	}
+
+	return count;
+}
+
+/**
+ * Creates a new parser based over a given  buffer with an array of tokens
+ * available.
+ */
+void jsmn_init(jsmn_parser *parser) {
+	parser->pos = 0;
+	parser->toknext = 0;
+	parser->toksuper = -1;
+}
+

data/ext/fast_mmaped_file/mmap.c

@@ -1,11 +1,11 @@
+#include "mmap.h"
+
 #include <errno.h>
 #include <fcntl.h>
-#include <ruby.h>
 #include <ruby/util.h>
 #include <sys/mman.h>
 
 #include "file_format.h"
-#include "mmap.h"
 #include "utils.h"
 
 #if 0
@@ -42,7 +42,6 @@ static VALUE mm_str(VALUE obj, int modify) {
     GET_MMAP(obj, i_mm, modify & ~MM_ORIGIN);
     if (modify & MM_MODIFY) {
         if (i_mm->t->flag & MM_FROZEN) rb_error_frozen("mmap");
-        if (!OBJ_TAINTED(ret) && rb_safe_level() >= 4) rb_raise(rb_eSecurityError, "Insecure: can't modify mmap");
     }
     ret = rb_obj_alloc(rb_cString);
     if (rb_obj_tainted(obj)) {
@@ -170,6 +169,35 @@ VALUE mm_s_alloc(VALUE obj) {
     return res;
 }
 
+size_t next_page_boundary(size_t value) {
+    size_t page_size = sysconf(_SC_PAGESIZE);
+
+    while (page_size < value) {
+        page_size *= 2;
+    }
+
+    return page_size;
+}
+
+/* Reference implementations:
+ * mozilla: https://hg.mozilla.org/mozilla-central/file/3d846420a907/xpcom/glue/FileUtils.cpp#l71
+ * glibc: https://github.com/lattera/glibc/blob/master/sysdeps/posix/posix_fallocate.c
+ */
+int reserve_mmap_file_bytes(int fd, size_t size) {
+#if __linux__
+    /* From https://stackoverflow.com/a/22820221: The difference with
+     * ftruncate(2) is that (on file systems supporting it, e.g. Ext4)
+     * disk space is indeed reserved by posix_fallocate but ftruncate
+     * extends the file by adding holes (and without reserving disk
+     * space). */
+    return posix_fallocate(fd, 0, size);
+#else
+    /* We simplify the reference implemnetations since we generally
+     * don't need to reserve more than a page size. */
+    return ftruncate(fd, size);
+#endif
+}
+
 VALUE mm_init(VALUE obj, VALUE fname) {
     struct stat st;
     int fd, smode = 0, pmode = 0, vscope, perm, init;
@@ -187,13 +215,6 @@ VALUE mm_init(VALUE obj, VALUE fname) {
     SafeStringValue(fname);
     path = StringValuePtr(fname);
 
-    {
-        if (rb_safe_level() > 0 && OBJ_TAINTED(fname)) {
-            rb_raise(rb_eSecurityError, "Insecure operation");
-        }
-        rb_secure(1);
-    }
-
     vscope = MAP_SHARED;
     size = 0;
     perm = 0666;
@@ -206,6 +227,7 @@ VALUE mm_init(VALUE obj, VALUE fname) {
     }
 
     if (fstat(fd, &st) == -1) {
+        close(fd);
         rb_raise(rb_eArgError, "Can't stat %s", path);
     }
     size = st.st_size;
@@ -215,17 +237,21 @@ VALUE mm_init(VALUE obj, VALUE fname) {
     offset = 0;
     init = 0;
 
-    if (size == 0 && (smode & O_RDWR)) {
-        if (lseek(fd, INITIAL_SIZE - 1, SEEK_END) == -1) {
-            rb_raise(rb_eIOError, "Can't lseek %zu", INITIAL_SIZE - 1);
-        }
-        if (write(fd, "\000", 1) != 1) {
-            rb_raise(rb_eIOError, "Can't extend %s", path);
-        }
+    if (size == 0) {
         init = 1;
         size = INITIAL_SIZE;
     }
 
+    /* We need to ensure the underlying file descriptor is at least a page size.
+     * Otherwise, we could get a SIGBUS error if mmap() attempts to read or write
+     * past the file. */
+    size_t reserve_size = next_page_boundary(size);
+
+    if (reserve_mmap_file_bytes(fd, reserve_size) != 0) {
+        close(fd);
+        rb_raise(rb_eIOError, "Can't reserve %zu bytes for memory-mapped file in %s", reserve_size, path);
+    }
+
     addr = mmap(0, size, pmode, vscope, fd, offset);
 
     if (addr == MAP_FAILED || !addr) {

data/ext/fast_mmaped_file/rendering.c

@@ -1,9 +1,10 @@
+#include "rendering.h"
+
 #include <float.h>
 #include <jsmn.h>
 
 #include "file_parsing.h"
 #include "globals.h"
-#include "rendering.h"
 #include "utils.h"
 
 #ifndef DBL_DECIMAL_DIG

data/ext/fast_mmaped_file/utils.c

@@ -1,8 +1,7 @@
-#include <errno.h>
-#include <ruby.h>
-
 #include "utils.h"
 
+#include <errno.h>
+
 static void rb_save_exception(VALUE exception, VALUE message) {
     VALUE current_thread = rb_thread_current();
 

data/ext/fast_mmaped_file/utils.h

@@ -1,5 +1,6 @@
 #ifndef UNUSED_H
 #define UNUSED_H
+#include <ruby.h>
 
 #ifdef UNUSED
 #elif defined(__GNUC__)

data/ext/fast_mmaped_file/value_access.c

@@ -1,15 +1,13 @@
-#include <ruby.h>
-#include <ruby/intern.h>
+#include "value_access.h"
 
 #include <errno.h>
 #include <fcntl.h>
+#include <ruby/intern.h>
 #include <sys/mman.h>
 #include <unistd.h>
 
 #include "file_format.h"
 #include "mmap.h"
-#include "value_access.h"
-
 #include "utils.h"
 
 static void close_file(mm_ipc *i_mm) {

data/ext/fast_mmaped_file/value_access.h

@@ -1,5 +1,6 @@
 #ifndef VALUE_ACCESS_H
 #define VALUE_ACCESS_H
+#include <ruby.h>
 
 VALUE method_load_used(VALUE self);
 

data/lib/prometheus/client.rb

@@ -29,13 +29,25 @@ module Prometheus
         configuration.pid_provider.call
       end
 
+      # Resets the registry and reinitializes all metrics files.
+      # Use case: clean up everything in specs `before` block,
+      # to prevent leaking the state between specs which are updating metrics.
       def reset!
         @registry = nil
         ::Prometheus::Client::MmapedValue.reset_and_reinitialize
       end
 
-      def reinitialize_on_pid_change
-        ::Prometheus::Client::MmapedValue.reinitialize_on_pid_change
+      # With `force: false`: reinitializes metric files only for processes with the changed PID.
+      # With `force: true`: reinitializes all metrics files.
+      # Always keeps the registry.
+      # Use case  (`force: false`): pick up new metric files on each worker start,
+      # without resetting already registered files for the master or previously initialized workers.
+      def reinitialize_on_pid_change(force: false)
+        if force
+          ::Prometheus::Client::MmapedValue.reset_and_reinitialize
+        else
+          ::Prometheus::Client::MmapedValue.reinitialize_on_pid_change
+        end
       end
     end
   end

data/lib/prometheus/client/label_set_validator.rb

@@ -19,14 +19,15 @@ module Prometheus
       end
 
       def valid?(labels)
-        unless labels.respond_to?(:all?)
+        unless labels.is_a?(Hash)
           raise InvalidLabelSetError, "#{labels} is not a valid label set"
         end
 
-        labels.all? do |key, _|
+        labels.all? do |key, value|
           validate_symbol(key)
           validate_name(key)
           validate_reserved_key(key)
+          validate_value(key, value)
         end
       end
 
@@ -65,6 +66,17 @@ module Prometheus
 
         raise ReservedLabelError, "#{key} is reserved"
       end
+
+      def validate_value(key, value)
+        return true if value.is_a?(String) ||
+                       value.is_a?(Numeric) ||
+                       value.is_a?(Symbol) ||
+                       value.is_a?(FalseClass) ||
+                       value.is_a?(TrueClass) ||
+                       value.nil?
+
+        raise InvalidLabelError, "#{key} does not contain a valid value (type #{value.class})"
+      end
     end
   end
 end

data/lib/prometheus/client/mmaped_dict.rb

@@ -57,6 +57,10 @@ module Prometheus
         Prometheus::Client.logger.warn("munmap raised error #{e}")
       end
 
+      def inspect
+        "#<#{self.class}:0x#{(object_id << 1).to_s(16)}>"
+      end
+
       private
 
       def init_value(key)

data/lib/prometheus/client/version.rb

@@ -1,5 +1,5 @@
 module Prometheus
   module Client
-    VERSION = '0.9.8'.freeze
+    VERSION = '0.12.0'.freeze
   end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: prometheus-client-mmap
 version: !ruby/object:Gem::Version
-  version: 0.9.8
+  version: 0.12.0
 platform: ruby
 authors:
 - Tobias Schmidt
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2019-07-08 00:00:00.000000000 Z
+date: 2020-09-22 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: fuzzbert
@@ -106,6 +106,8 @@ files:
 - ext/fast_mmaped_file/file_reading.c
 - ext/fast_mmaped_file/file_reading.h
 - ext/fast_mmaped_file/globals.h
+- ext/fast_mmaped_file/hashmap.c
+- ext/fast_mmaped_file/jsmn.c
 - ext/fast_mmaped_file/mmap.c
 - ext/fast_mmaped_file/mmap.h
 - ext/fast_mmaped_file/rendering.c
@@ -114,6 +116,7 @@ files:
 - ext/fast_mmaped_file/utils.h
 - ext/fast_mmaped_file/value_access.c
 - ext/fast_mmaped_file/value_access.h
+- lib/fast_mmaped_file.bundle
 - lib/prometheus.rb
 - lib/prometheus/client.rb
 - lib/prometheus/client/configuration.rb