1 files changed, 381 insertions, 294 deletions

diff --git a/lib/libalpm/util.c b/lib/libalpm/util.c
index 0e5e8a00..34f817e1 100644
--- a/lib/libalpm/util.c
+++ b/lib/libalpm/util.c
@@ -22,24 +22,17 @@
  *  along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
-#include "config.h"
-
-#include <stdio.h>
 #include <stdlib.h>
-#include <stdarg.h>
-#include <string.h>
 #include <unistd.h>
 #include <ctype.h>
 #include <dirent.h>
-#include <fcntl.h>
 #include <time.h>
 #include <syslog.h>
 #include <errno.h>
 #include <limits.h>
-#include <sys/types.h>
-#include <sys/stat.h>
 #include <sys/wait.h>
 #include <locale.h> /* setlocale */
+#include <fnmatch.h>
 
 /* libarchive */
 #include <archive.h>
@@ -62,8 +55,16 @@
 #include "trans.h"
 
 #ifndef HAVE_STRSEP
-/* This is a replacement for strsep which is not portable (missing on Solaris).
- * Copyright (c) 2001 by François Gouget <fgouget_at_codeweavers.com> */
+/** Extracts tokens from a string.
+ * Replaces strset which is not portable (missing on Solaris).
+ * Copyright (c) 2001 by François Gouget <fgouget_at_codeweavers.com>
+ * Modifies str to point to the first character after the token if one is
+ * found, or NULL if one is not.
+ * @param str string containing delimited tokens to parse
+ * @param delim character delimiting tokens in str
+ * @return pointer to the first token in str if str is not NULL, NULL if
+ * str is NULL
+ */
 char* strsep(char** str, const char* delims)
 {
 	char* token;
@@ -93,128 +94,106 @@ int _alpm_makepath(const char *path)
 	return _alpm_makepath_mode(path, 0755);
 }
 
-/* does the same thing as 'mkdir -p' */
+/** Creates a directory, including parents if needed, similar to 'mkdir -p'.
+ * @param path directory path to create
+ * @param mode permission mode for created directories
+ * @return 0 on success, 1 on error
+ */
 int _alpm_makepath_mode(const char *path, mode_t mode)
 {
-	/* A bit of pointer hell here. Descriptions:
-	 * orig - a copy of path so we can safely butcher it with strsep
-	 * str - the current position in the path string (after the delimiter)
-	 * ptr - the original position of str after calling strsep
-	 * incr - incrementally generated path for use in stat/mkdir call
-	 */
-	char *orig, *str, *ptr, *incr;
-	mode_t oldmask = umask(0000);
+	char *ptr, *str;
+	mode_t oldmask;
 	int ret = 0;
 
-	orig = strdup(path);
-	incr = calloc(strlen(orig) + 1, sizeof(char));
-	str = orig;
-	while((ptr = strsep(&str, "/"))) {
-		if(strlen(ptr)) {
-			/* we have another path component- append the newest component to
-			 * existing string and create one more level of dir structure */
-			strcat(incr, "/");
-			strcat(incr, ptr);
-			if(access(incr, F_OK)) {
-				if(mkdir(incr, mode)) {
-					ret = 1;
-					break;
-				}
-			}
-		}
-	}
-	free(orig);
-	free(incr);
-	umask(oldmask);
-	return ret;
-}
+	STRDUP(str, path, return 1);
 
-int _alpm_copyfile(const char *src, const char *dest)
-{
-	FILE *in, *out;
-	size_t len;
-	char *buf;
-	int ret = 0;
+	oldmask = umask(0000);
 
-	in = fopen(src, "rb");
-	if(in == NULL) {
-		return 1;
-	}
-	out = fopen(dest, "wb");
-	if(out == NULL) {
-		fclose(in);
-		return 1;
-	}
+	for(ptr = str; *ptr; ptr++) {
+		/* detect mid-path condition and zero length paths */
+		if(*ptr != '/' || ptr == str || ptr[-1] == '/') {
+			continue;
+		}
 
-	MALLOC(buf, (size_t)ALPM_BUFFER_SIZE, ret = 1; goto cleanup);
+		/* temporarily mask the end of the path */
+		*ptr = '\0';
 
-	/* do the actual file copy */
-	while((len = fread(buf, 1, ALPM_BUFFER_SIZE, in))) {
-		size_t nwritten = 0;
-		nwritten = fwrite(buf, 1, len, out);
-		if((nwritten != len) || ferror(out)) {
-			ret = -1;
-			goto cleanup;
+		if(mkdir(str, 0755) < 0 && errno != EEXIST) {
+			ret = 1;
+			goto done;
 		}
+
+		/* restore path separator */
+		*ptr = '/';
 	}
 
-	/* chmod dest to permissions of src, as long as it is not a symlink */
-	struct stat statbuf;
-	if(!stat(src, &statbuf)) {
-		if(! S_ISLNK(statbuf.st_mode)) {
-			fchmod(fileno(out), statbuf.st_mode);
-		}
-	} else {
-		/* stat was unsuccessful */
+	/* end of the string. add the full path. It will already exist when the path
+	 * passed in has a trailing slash. */
+	if(mkdir(str, 0755) < 0 && errno != EEXIST) {
 		ret = 1;
 	}
 
-cleanup:
-	fclose(in);
-	fclose(out);
-	free(buf);
+done:
+	umask(oldmask);
+	free(str);
 	return ret;
 }
 
-/* Trim whitespace and newlines from a string
-*/
-char *_alpm_strtrim(char *str)
+/** Copies a file.
+ * @param src file path to copy from
+ * @param dest file path to copy to
+ * @return 0 on success, 1 on error 
+ */
+int _alpm_copyfile(const char *src, const char *dest)
 {
-	char *pch = str;
+	char *buf;
+	int in, out, ret = 1;
+	ssize_t nread;
+	struct stat st;
 
-	if(*str == '\0') {
-		/* string is empty, so we're done. */
-		return str;
+	MALLOC(buf, (size_t)ALPM_BUFFER_SIZE, return 1);
+
+	OPEN(in, src, O_RDONLY);
+	do {
+		out = open(dest, O_WRONLY | O_CREAT, 0000);
+	} while(out == -1 && errno == EINTR);
+	if(in < 0 || out < 0) {
+		goto cleanup;
 	}
 
-	while(isspace((unsigned char)*pch)) {
-		pch++;
+	if(fstat(in, &st) || fchmod(out, st.st_mode)) {
+		goto cleanup;
 	}
-	if(pch != str) {
-		size_t len = strlen(pch);
-		if(len) {
-			memmove(str, pch, len + 1);
-		} else {
-			*str = '\0';
+
+	/* do the actual file copy */
+	while((nread = read(in, buf, ALPM_BUFFER_SIZE)) > 0 || errno == EINTR) {
+		ssize_t nwrite = 0;
+		if(nread < 0) {
+			continue;
 		}
+		do {
+			nwrite = write(out, buf + nwrite, nread);
+			if(nwrite >= 0) {
+				nread -= nwrite;
+			} else if(errno != EINTR) {
+				goto cleanup;
+			}
+		} while(nread > 0);
 	}
+	ret = 0;
 
-	/* check if there wasn't anything but whitespace in the string. */
-	if(*str == '\0') {
-		return str;
+cleanup:
+	free(buf);
+	if(in >= 0) {
+		CLOSE(in);
 	}
-
-	pch = (str + (strlen(str) - 1));
-	while(isspace((unsigned char)*pch)) {
-		pch--;
+	if(out >= 0) {
+		CLOSE(out);
 	}
-	*++pch = '\0';
-
-	return str;
+	return ret;
 }
 
-/**
- * Trim trailing newline from a string (if one exists).
+/** Trim trailing newlines from a string (if any exist).
  * @param str a single line of text
  * @return the length of the trimmed string
  */
@@ -235,9 +214,70 @@ size_t _alpm_strip_newline(char *str)
 
 /* Compression functions */
 
-/**
- * @brief Unpack a specific file in an archive.
- *
+/** Open an archive for reading and perform the necessary boilerplate.
+ * This takes care of creating the libarchive 'archive' struct, setting up
+ * compression and format options, opening a file descriptor, setting up the
+ * buffer size, and performing a stat on the path once opened.
+ * On error, no file descriptor is opened, and the archive pointer returned
+ * will be set to NULL.
+ * @param handle the context handle
+ * @param path the path of the archive to open
+ * @param buf space for a stat buffer for the given path
+ * @param archive pointer to place the created archive object
+ * @param error error code to set on failure to open archive
+ * @return -1 on failure, >=0 file descriptor on success
+ */
+int _alpm_open_archive(alpm_handle_t *handle, const char *path,
+		struct stat *buf, struct archive **archive, alpm_errno_t error)
+{
+	int fd;
+	size_t bufsize = ALPM_BUFFER_SIZE;
+	errno = 0;
+
+	if((*archive = archive_read_new()) == NULL) {
+		RET_ERR(handle, ALPM_ERR_LIBARCHIVE, -1);
+	}
+
+	archive_read_support_compression_all(*archive);
+	archive_read_support_format_all(*archive);
+
+	_alpm_log(handle, ALPM_LOG_DEBUG, "opening archive %s\n", path);
+	OPEN(fd, path, O_RDONLY);
+	if(fd < 0) {
+		_alpm_log(handle, ALPM_LOG_ERROR,
+				_("could not open file %s: %s\n"), path, strerror(errno));
+		goto error;
+	}
+
+	if(fstat(fd, buf) != 0) {
+		_alpm_log(handle, ALPM_LOG_ERROR,
+				_("could not stat file %s: %s\n"), path, strerror(errno));
+		goto error;
+	}
+#ifdef HAVE_STRUCT_STAT_ST_BLKSIZE
+	if(buf->st_blksize > ALPM_BUFFER_SIZE) {
+		bufsize = buf->st_blksize;
+	}
+#endif
+
+	if(archive_read_open_fd(*archive, fd, bufsize) != ARCHIVE_OK) {
+		_alpm_log(handle, ALPM_LOG_ERROR, _("could not open file %s: %s\n"),
+				path, archive_error_string(*archive));
+		goto error;
+	}
+
+	return fd;
+
+error:
+	archive_read_finish(*archive);
+	*archive = NULL;
+	if(fd >= 0) {
+		CLOSE(fd);
+	}
+	RET_ERR(handle, error, -1);
+}
+
+/** Unpack a specific file in an archive.
  * @param handle the context handle
  * @param archive the archive to unpack
  * @param prefix where to extract the files
@@ -258,46 +298,33 @@ int _alpm_unpack_single(alpm_handle_t *handle, const char *archive,
 	return ret;
 }
 
-/**
- * @brief Unpack a list of files in an archive.
- *
+/** Unpack a list of files in an archive.
  * @param handle the context handle
- * @param archive the archive to unpack
+ * @param path the archive to unpack
  * @param prefix where to extract the files
  * @param list a list of files within the archive to unpack or NULL for all
  * @param breakfirst break after the first entry found
- *
  * @return 0 on success, 1 on failure
  */
-int _alpm_unpack(alpm_handle_t *handle, const char *archive, const char *prefix,
+int _alpm_unpack(alpm_handle_t *handle, const char *path, const char *prefix,
 		alpm_list_t *list, int breakfirst)
 {
 	int ret = 0;
 	mode_t oldmask;
-	struct archive *_archive;
+	struct archive *archive;
 	struct archive_entry *entry;
-	int cwdfd;
-
-	if((_archive = archive_read_new()) == NULL) {
-		RET_ERR(handle, ALPM_ERR_LIBARCHIVE, 1);
-	}
-
-	archive_read_support_compression_all(_archive);
-	archive_read_support_format_all(_archive);
+	struct stat buf;
+	int fd, cwdfd;
 
-	if(archive_read_open_filename(_archive, archive,
-				ALPM_BUFFER_SIZE) != ARCHIVE_OK) {
-		_alpm_log(handle, ALPM_LOG_ERROR, _("could not open file %s: %s\n"), archive,
-				archive_error_string(_archive));
-		RET_ERR(handle, ALPM_ERR_PKG_OPEN, 1);
+	fd = _alpm_open_archive(handle, path, &buf, &archive, ALPM_ERR_PKG_OPEN);
+	if(fd < 0) {
+		return 1;
 	}
 
 	oldmask = umask(0022);
 
 	/* save the cwd so we can restore it later */
-	do {
-		cwdfd = open(".", O_RDONLY);
-	} while(cwdfd == -1 && errno == EINTR);
+	OPEN(cwdfd, ".", O_RDONLY);
 	if(cwdfd < 0) {
 		_alpm_log(handle, ALPM_LOG_ERROR, _("could not get current working directory\n"));
 	}
@@ -310,19 +337,12 @@ int _alpm_unpack(alpm_handle_t *handle, const char *archive, const char *prefix,
 		goto cleanup;
 	}
 
-	while(archive_read_next_header(_archive, &entry) == ARCHIVE_OK) {
-		const struct stat *st;
-		const char *entryname; /* the name of the file in the archive */
+	while(archive_read_next_header(archive, &entry) == ARCHIVE_OK) {
+		const char *entryname;
+		mode_t mode;
 
-		st = archive_entry_stat(entry);
 		entryname = archive_entry_pathname(entry);
 
-		if(S_ISREG(st->st_mode)) {
-			archive_entry_set_perm(entry, 0644);
-		} else if(S_ISDIR(st->st_mode)) {
-			archive_entry_set_perm(entry, 0755);
-		}
-
 		/* If specific files were requested, skip entries that don't match. */
 		if(list) {
 			char *entry_prefix = strdup(entryname);
@@ -333,7 +353,7 @@ int _alpm_unpack(alpm_handle_t *handle, const char *archive, const char *prefix,
 			char *found = alpm_list_find_str(list, entry_prefix);
 			free(entry_prefix);
 			if(!found) {
-				if(archive_read_data_skip(_archive) != ARCHIVE_OK) {
+				if(archive_read_data_skip(archive) != ARCHIVE_OK) {
 					ret = 1;
 					goto cleanup;
 				}
@@ -343,15 +363,22 @@ int _alpm_unpack(alpm_handle_t *handle, const char *archive, const char *prefix,
 			}
 		}
 
+		mode = archive_entry_mode(entry);
+		if(S_ISREG(mode)) {
+			archive_entry_set_perm(entry, 0644);
+		} else if(S_ISDIR(mode)) {
+			archive_entry_set_perm(entry, 0755);
+		}
+
 		/* Extract the archive entry. */
-		int readret = archive_read_extract(_archive, entry, 0);
+		int readret = archive_read_extract(archive, entry, 0);
 		if(readret == ARCHIVE_WARN) {
 			/* operation succeeded but a non-critical error was encountered */
 			_alpm_log(handle, ALPM_LOG_WARNING, _("warning given when extracting %s (%s)\n"),
-					entryname, archive_error_string(_archive));
+					entryname, archive_error_string(archive));
 		} else if(readret != ARCHIVE_OK) {
 			_alpm_log(handle, ALPM_LOG_ERROR, _("could not extract %s (%s)\n"),
-					entryname, archive_error_string(_archive));
+					entryname, archive_error_string(archive));
 			ret = 1;
 			goto cleanup;
 		}
@@ -363,63 +390,20 @@ int _alpm_unpack(alpm_handle_t *handle, const char *archive, const char *prefix,
 
 cleanup:
 	umask(oldmask);
-	archive_read_finish(_archive);
+	archive_read_finish(archive);
+	CLOSE(fd);
 	if(cwdfd >= 0) {
 		if(fchdir(cwdfd) != 0) {
 			_alpm_log(handle, ALPM_LOG_ERROR,
 					_("could not restore working directory (%s)\n"), strerror(errno));
 		}
-		close(cwdfd);
+		CLOSE(cwdfd);
 	}
 
 	return ret;
 }
 
-/* does the same thing as 'rm -rf' */
-int _alpm_rmrf(const char *path)
-{
-	int errflag = 0;
-	struct dirent *dp;
-	DIR *dirp;
-	struct stat st;
-
-	if(_alpm_lstat(path, &st) == 0) {
-		if(!S_ISDIR(st.st_mode)) {
-			if(!unlink(path)) {
-				return 0;
-			} else {
-				if(errno == ENOENT) {
-					return 0;
-				} else {
-					return 1;
-				}
-			}
-		} else {
-			dirp = opendir(path);
-			if(!dirp) {
-				return 1;
-			}
-			for(dp = readdir(dirp); dp != NULL; dp = readdir(dirp)) {
-				if(dp->d_name) {
-					if(strcmp(dp->d_name, "..") != 0 && strcmp(dp->d_name, ".") != 0) {
-						char name[PATH_MAX];
-						sprintf(name, "%s/%s", path, dp->d_name);
-						errflag += _alpm_rmrf(name);
-					}
-				}
-			}
-			closedir(dirp);
-			if(rmdir(path)) {
-				errflag++;
-			}
-		}
-		return errflag;
-	}
-	return 0;
-}
-
-/**
- * Determine if there are files in a directory
+/** Determine if there are files in a directory.
  * @param handle the context handle
  * @param path the full absolute directory path
  * @param full_count whether to return an exact count of files
@@ -460,6 +444,13 @@ ssize_t _alpm_files_in_directory(alpm_handle_t *handle, const char *path,
 	return files;
 }
 
+/** Write formatted message to log.
+ * @param handle the context handle
+ * @param format formatted string to write out
+ * @param args formatting arguments
+ * @return 0 or number of characters written on success, vfprintf return value
+ * on error
+ */
 int _alpm_logaction(alpm_handle_t *handle, const char *fmt, va_list args)
 {
 	int ret = 0;
@@ -491,16 +482,20 @@ int _alpm_logaction(alpm_handle_t *handle, const char *fmt, va_list args)
 	return ret;
 }
 
-int _alpm_run_chroot(alpm_handle_t *handle, const char *path, char *const argv[])
+/** Execute a command with arguments in a chroot.
+ * @param handle the context handle
+ * @param cmd command to execute
+ * @param argv arguments to pass to cmd
+ * @return 0 on success, 1 on error
+ */
+int _alpm_run_chroot(alpm_handle_t *handle, const char *cmd, char *const argv[])
 {
 	pid_t pid;
 	int pipefd[2], cwdfd;
 	int retval = 0;
 
 	/* save the cwd so we can restore it later */
-	do {
-		cwdfd = open(".", O_RDONLY);
-	} while(cwdfd == -1 && errno == EINTR);
+	OPEN(cwdfd, ".", O_RDONLY);
 	if(cwdfd < 0) {
 		_alpm_log(handle, ALPM_LOG_ERROR, _("could not get current working directory\n"));
 	}
@@ -513,7 +508,7 @@ int _alpm_run_chroot(alpm_handle_t *handle, const char *path, char *const argv[]
 	}
 
 	_alpm_log(handle, ALPM_LOG_DEBUG, "executing \"%s\" under chroot \"%s\"\n",
-			path, handle->root);
+			cmd, handle->root);
 
 	/* Flush open fds before fork() to avoid cloning buffers */
 	fflush(NULL);
@@ -534,12 +529,12 @@ int _alpm_run_chroot(alpm_handle_t *handle, const char *path, char *const argv[]
 
 	if(pid == 0) {
 		/* this code runs for the child only (the actual chroot/exec) */
-		close(1);
-		close(2);
+		CLOSE(1);
+		CLOSE(2);
 		while(dup2(pipefd[1], 1) == -1 && errno == EINTR);
 		while(dup2(pipefd[1], 2) == -1 && errno == EINTR);
-		close(pipefd[0]);
-		close(pipefd[1]);
+		CLOSE(pipefd[0]);
+		CLOSE(pipefd[1]);
 
 		/* use fprintf instead of _alpm_log to send output through the parent */
 		if(chroot(handle->root) != 0) {
@@ -552,7 +547,8 @@ int _alpm_run_chroot(alpm_handle_t *handle, const char *path, char *const argv[]
 			exit(1);
 		}
 		umask(0022);
-		execv(path, argv);
+		execv(cmd, argv);
+		/* execv only returns if there was an error */
 		fprintf(stderr, _("call to execv failed (%s)\n"), strerror(errno));
 		exit(1);
 	} else {
@@ -560,10 +556,10 @@ int _alpm_run_chroot(alpm_handle_t *handle, const char *path, char *const argv[]
 		int status;
 		FILE *pipe_file;
 
-		close(pipefd[1]);
+		CLOSE(pipefd[1]);
 		pipe_file = fdopen(pipefd[0], "r");
 		if(pipe_file == NULL) {
-			close(pipefd[0]);
+			CLOSE(pipefd[0]);
 			retval = 1;
 		} else {
 			while(!feof(pipe_file)) {
@@ -605,12 +601,16 @@ cleanup:
 			_alpm_log(handle, ALPM_LOG_ERROR,
 					_("could not restore working directory (%s)\n"), strerror(errno));
 		}
-		close(cwdfd);
+		CLOSE(cwdfd);
 	}
 
 	return retval;
 }
 
+/** Run ldconfig in a chroot.
+ * @param handle the context handle
+ * @return 0 on success, 1 on error
+ */
 int _alpm_ldconfig(alpm_handle_t *handle)
 {
 	char line[PATH_MAX];
@@ -629,8 +629,13 @@ int _alpm_ldconfig(alpm_handle_t *handle)
 	return 0;
 }
 
-/* Helper function for comparing strings using the
- * alpm "compare func" signature */
+/** Helper function for comparing strings using the alpm "compare func"
+ * signature.
+ * @param s1 first string to be compared
+ * @param s2 second string to be compared
+ * @return 0 if strings are equal, positive int if first unequal character
+ * has a greater value in s1, negative if it has a greater value in s2
+ */
 int _alpm_str_cmp(const void *s1, const void *s2)
 {
 	return strcmp(s1, s2);
@@ -687,7 +692,7 @@ const char *_alpm_filecache_setup(alpm_handle_t *handle)
 		} else if(!S_ISDIR(buf.st_mode)) {
 			_alpm_log(handle, ALPM_LOG_DEBUG,
 					"skipping cachedir, not a directory: %s\n", cachedir);
-		} else if(access(cachedir, W_OK) != 0) {
+		} else if(_alpm_access(handle, NULL, cachedir, W_OK) != 0) {
 			_alpm_log(handle, ALPM_LOG_DEBUG,
 					"skipping cachedir, not writable: %s\n", cachedir);
 		} else if(!(buf.st_mode & (S_IWUSR | S_IWGRP | S_IWOTH))) {
@@ -739,51 +744,64 @@ int _alpm_lstat(const char *path, struct stat *buf)
 }
 
 #ifdef HAVE_LIBSSL
+/** Compute the MD5 message digest of a file.
+ * @param path file path of file to compute  MD5 digest of
+ * @param output string to hold computed MD5 digest
+ * @return 0 on success, 1 on file open error, 2 on file read error
+ */
 static int md5_file(const char *path, unsigned char output[16])
 {
-	FILE *f;
-	size_t n;
 	MD5_CTX ctx;
 	unsigned char *buf;
+	ssize_t n;
+	int fd;
 
-	CALLOC(buf, ALPM_BUFFER_SIZE, sizeof(unsigned char), return 1);
+	MALLOC(buf, (size_t)ALPM_BUFFER_SIZE, return 1);
 
-	if((f = fopen(path, "rb")) == NULL) {
+	OPEN(fd, path, O_RDONLY);
+	if(fd < 0) {
 		free(buf);
 		return 1;
 	}
 
 	MD5_Init(&ctx);
 
-	while((n = fread(buf, 1, ALPM_BUFFER_SIZE, f)) > 0) {
+	while((n = read(fd, buf, ALPM_BUFFER_SIZE)) > 0 || errno == EINTR) {
+		if(n < 0) {
+			continue;
+		}
 		MD5_Update(&ctx, buf, n);
 	}
 
-	MD5_Final(output, &ctx);
-
-	memset(&ctx, 0, sizeof(MD5_CTX));
+	CLOSE(fd);
 	free(buf);
 
-	if(ferror(f) != 0) {
-		fclose(f);
+	if(n < 0) {
 		return 2;
 	}
 
-	fclose(f);
+	MD5_Final(output, &ctx);
 	return 0;
 }
 
 /* third param is so we match the PolarSSL definition */
+/** Compute the SHA-224 or SHA-256 message digest of a file.
+ * @param path file path of file to compute SHA2 digest of
+ * @param output string to hold computed SHA2 digest
+ * @param is224 use SHA-224 instead of SHA-256
+ * @return 0 on success, 1 on file open error, 2 on file read error
+ */
 static int sha2_file(const char *path, unsigned char output[32], int is224)
 {
-	FILE *f;
-	size_t n;
 	SHA256_CTX ctx;
 	unsigned char *buf;
+	ssize_t n;
+	int fd;
 
-	CALLOC(buf, ALPM_BUFFER_SIZE, sizeof(unsigned char), return 1);
+	MALLOC(buf, (size_t)ALPM_BUFFER_SIZE, return 1);
 
-	if((f = fopen(path, "rb")) == NULL) {
+	OPEN(fd, path, O_RDONLY);
+	if(fd < 0) {
 		free(buf);
 		return 1;
 	}
@@ -794,7 +812,10 @@ static int sha2_file(const char *path, unsigned char output[32], int is224)
 		SHA256_Init(&ctx);
 	}
 
-	while((n = fread(buf, 1, ALPM_BUFFER_SIZE, f)) > 0) {
+	while((n = read(fd, buf, ALPM_BUFFER_SIZE)) > 0 || errno == EINTR) {
+		if(n < 0) {
+			continue;
+		}
 		if(is224) {
 			SHA224_Update(&ctx, buf, n);
 		} else {
@@ -802,24 +823,45 @@ static int sha2_file(const char *path, unsigned char output[32], int is224)
 		}
 	}
 
+	CLOSE(fd);
+	free(buf);
+
+	if(n < 0) {
+		return 2;
+	}
+
 	if(is224) {
 		SHA224_Final(output, &ctx);
 	} else {
 		SHA256_Final(output, &ctx);
 	}
+	return 0;
+}
+#endif
 
-	memset(&ctx, 0, sizeof(SHA256_CTX));
-	free(buf);
+/** Create a string representing bytes in hexadecimal.
+ * @param bytes the bytes to represent in hexadecimal
+ * @param size number of bytes to consider
+ * @return a NULL terminated string with the hexadecimal representation of
+ * bytes or NULL on error. This string must be freed.
+ */
+static char *hex_representation(unsigned char *bytes, size_t size)
+{
+	static const char *hex_digits = "0123456789abcdef";
+	char *str;
+	size_t i;
 
-	if(ferror(f) != 0) {
-		fclose(f);
-		return 2;
+	MALLOC(str, 2 * size + 1, return NULL);
+
+	for (i = 0; i < size; i++) {
+		str[2 * i] = hex_digits[bytes[i] >> 4];
+		str[2 * i + 1] = hex_digits[bytes[i] & 0x0f];
 	}
 
-	fclose(f);
-	return 0;
+	str[2 * size] = '\0';
+
+	return str;
 }
-#endif
 
 /** Get the md5 sum of file.
  * @param filename name of the file
@@ -829,28 +871,15 @@ static int sha2_file(const char *path, unsigned char output[32], int is224)
 char SYMEXPORT *alpm_compute_md5sum(const char *filename)
 {
 	unsigned char output[16];
-	char *md5sum;
-	int ret, i;
 
 	ASSERT(filename != NULL, return NULL);
 
-	/* allocate 32 chars plus 1 for null */
-	CALLOC(md5sum, 33, sizeof(char), return NULL);
 	/* defined above for OpenSSL, otherwise defined in md5.h */
-	ret = md5_file(filename, output);
-
-	if(ret > 0) {
-		free(md5sum);
+	if(md5_file(filename, output) > 0) {
 		return NULL;
 	}
 
-	/* Convert the result to something readable */
-	for (i = 0; i < 16; i++) {
-		/* sprintf is acceptable here because we know our output */
-		sprintf(md5sum +(i * 2), "%02x", output[i]);
-	}
-
-	return md5sum;
+	return hex_representation(output, 16);
 }
 
 /** Get the sha256 sum of file.
@@ -861,30 +890,24 @@ char SYMEXPORT *alpm_compute_md5sum(const char *filename)
 char SYMEXPORT *alpm_compute_sha256sum(const char *filename)
 {
 	unsigned char output[32];
-	char *sha256sum;
-	int ret, i;
 
 	ASSERT(filename != NULL, return NULL);
 
-	/* allocate 64 chars plus 1 for null */
-	CALLOC(sha256sum, 65, sizeof(char), return NULL);
 	/* defined above for OpenSSL, otherwise defined in sha2.h */
-	ret = sha2_file(filename, output, 0);
-
-	if(ret > 0) {
-		free(sha256sum);
+	if(sha2_file(filename, output, 0) > 0) {
 		return NULL;
 	}
 
-	/* Convert the result to something readable */
-	for (i = 0; i < 32; i++) {
-		/* sprintf is acceptable here because we know our output */
-		sprintf(sha256sum +(i * 2), "%02x", output[i]);
-	}
-
-	return sha256sum;
+	return hex_representation(output, 32);
 }
 
+/** Calculates a file's MD5 or SHA2 digest  and compares it to an expected value. 
+ * @param filepath path of the file to check
+ * @param expected hash value to compare against
+ * @param type digest type to use
+ * @return 0 if file matches the expected hash, 1 if they do not match, -1 on
+ * error
+ */
 int _alpm_test_checksum(const char *filepath, const char *expected,
 		enum _alpm_csum type)
 {
@@ -912,18 +935,24 @@ int _alpm_test_checksum(const char *filepath, const char *expected,
 }
 
 /* Note: does NOT handle sparse files on purpose for speed. */
+/** TODO.
+ * Does not handle sparse files on purpose for speed.
+ * @param a
+ * @param b
+ * @return 
+ */
 int _alpm_archive_fgets(struct archive *a, struct archive_read_buffer *b)
 {
-	char *i = NULL;
-	int64_t offset;
-	int done = 0;
-
 	/* ensure we start populating our line buffer at the beginning */
 	b->line_offset = b->line;
 
 	while(1) {
+		size_t block_remaining;
+		char *eol;
+
 		/* have we processed this entire block? */
 		if(b->block + b->block_size == b->block_offset) {
+			int64_t offset;
 			if(b->ret == ARCHIVE_EOF) {
 				/* reached end of archive on the last read, now we are out of data */
 				goto cleanup;
@@ -933,20 +962,20 @@ int _alpm_archive_fgets(struct archive *a, struct archive_read_buffer *b)
 			b->ret = archive_read_data_block(a, (void *)&b->block,
 					&b->block_size, &offset);
 			b->block_offset = b->block;
+			block_remaining = b->block_size;
 
 			/* error, cleanup */
 			if(b->ret < ARCHIVE_OK) {
 				goto cleanup;
 			}
+		} else {
+			block_remaining = b->block + b->block_size - b->block_offset;
 		}
 
-		/* loop through the block looking for EOL characters */
-		for(i = b->block_offset; i < (b->block + b->block_size); i++) {
-			/* check if read value was null or newline */
-			if(*i == '\0' || *i == '\n') {
-				done = 1;
-				break;
-			}
+		/* look through the block looking for EOL characters */
+		eol = memchr(b->block_offset, '\n', block_remaining);
+		if(!eol) {
+			eol = memchr(b->block_offset, '\0', block_remaining);
 		}
 
 		/* allocate our buffer, or ensure our existing one is big enough */
@@ -956,8 +985,10 @@ int _alpm_archive_fgets(struct archive *a, struct archive_read_buffer *b)
 			b->line_size = b->block_size + 1;
 			b->line_offset = b->line;
 		} else {
-			size_t needed = (size_t)((b->line_offset - b->line)
-					+ (i - b->block_offset) + 1);
+			/* note: we know eol > b->block_offset and b->line_offset > b->line,
+			 * so we know the result is unsigned and can fit in size_t */
+			size_t new = eol ? (size_t)(eol - b->block_offset) : block_remaining;
+			size_t needed = (size_t)((b->line_offset - b->line) + new + 1);
 			if(needed > b->max_line_size) {
 				b->ret = -ERANGE;
 				goto cleanup;
@@ -974,11 +1005,11 @@ int _alpm_archive_fgets(struct archive *a, struct archive_read_buffer *b)
 			}
 		}
 
-		if(done) {
-			size_t len = (size_t)(i - b->block_offset);
+		if(eol) {
+			size_t len = (size_t)(eol - b->block_offset);
 			memcpy(b->line_offset, b->block_offset, len);
 			b->line_offset[len] = '\0';
-			b->block_offset = ++i;
+			b->block_offset = eol + 1;
 			/* this is the main return point; from here you can read b->line */
 			return ARCHIVE_OK;
 		} else {
@@ -986,7 +1017,7 @@ int _alpm_archive_fgets(struct archive *a, struct archive_read_buffer *b)
 			size_t len = (size_t)(b->block + b->block_size - b->block_offset);
 			memcpy(b->line_offset, b->block_offset, len);
 			b->line_offset += len;
-			b->block_offset = i;
+			b->block_offset = b->block + b->block_size;
 			/* there was no new data, return what is left; saved ARCHIVE_EOF will be
 			 * returned on next call */
 			if(len == 0) {
@@ -1005,6 +1036,14 @@ cleanup:
 	}
 }
 
+/** Parse a full package specifier.
+ * @param target package specifier to parse, such as: "pacman-4.0.1-2",
+ * "pacman-4.01-2/", or "pacman-4.0.1-2/desc"
+ * @param name to hold package name
+ * @param version to hold package version
+ * @param name_hash to hold package name hash
+ * @return 0 on success, -1 on error
+ */
 int _alpm_splitname(const char *target, char **name, char **version,
 		unsigned long *name_hash)
 {
@@ -1057,8 +1096,7 @@ int _alpm_splitname(const char *target, char **name, char **version,
 	return 0;
 }
 
-/**
- * Hash the given string to an unsigned long value.
+/** Hash the given string to an unsigned long value.
  * This is the standard sdbm hashing algorithm.
  * @param str string to hash
  * @return the hash value of the given string
@@ -1072,12 +1110,17 @@ unsigned long _alpm_hash_sdbm(const char *str)
 		return hash;
 	}
 	while((c = *str++)) {
-		hash = c + (hash << 6) + (hash << 16) - hash;
+		hash = c + hash * 65599;
 	}
 
 	return hash;
 }
 
+/** Convert a string to a file offset.
+ * This parses bare positive integers only.
+ * @param line string to convert
+ * @return off_t on success, -1 on error
+ */
 off_t _alpm_strtoofft(const char *line)
 {
 	char *end;
@@ -1089,13 +1132,13 @@ off_t _alpm_strtoofft(const char *line)
 		return (off_t)-1;
 	}
 	result = strtoull(line, &end, 10);
-	if (result == 0 && end == line) {
+	if(result == 0 && end == line) {
 		/* line was not a number */
 		return (off_t)-1;
-	} else if (result == ULLONG_MAX && errno == ERANGE) {
+	} else if(result == ULLONG_MAX && errno == ERANGE) {
 		/* line does not fit in unsigned long long */
 		return (off_t)-1;
-	} else if (*end) {
+	} else if(*end) {
 		/* line began with a number but has junk left over at the end */
 		return (off_t)-1;
 	}
@@ -1103,8 +1146,16 @@ off_t _alpm_strtoofft(const char *line)
 	return (off_t)result;
 }
 
-time_t _alpm_parsedate(const char *line)
+/** Parses a date into an alpm_time_t struct.
+ * @param line date to parse
+ * @return time struct on success, 0 on error
+ */
+alpm_time_t _alpm_parsedate(const char *line)
 {
+	char *end;
+	long long result;
+	errno = 0;
+
 	if(isalpha((unsigned char)line[0])) {
 		/* initialize to null in case of failure */
 		struct tm tmp_tm;
@@ -1112,13 +1163,27 @@ time_t _alpm_parsedate(const char *line)
 		setlocale(LC_TIME, "C");
 		strptime(line, "%a %b %e %H:%M:%S %Y", &tmp_tm);
 		setlocale(LC_TIME, "");
-		return mktime(&tmp_tm);
+		return (alpm_time_t)mktime(&tmp_tm);
+	}
+
+	result = strtoll(line, &end, 10);
+	if(result == 0 && end == line) {
+		/* line was not a number */
+		errno = EINVAL;
+		return 0;
+	} else if(errno == ERANGE) {
+		/* line does not fit in long long */
+		return 0;
+	} else if(*end) {
+		/* line began with a number but has junk left over at the end */
+		errno = EINVAL;
+		return 0;
 	}
-	return (time_t)atol(line);
+
+	return (alpm_time_t)result;
 }
 
-/**
- * Wrapper around access() which takes a dir and file argument
+/** Wrapper around access() which takes a dir and file argument
  * separately and generates an appropriate error message.
  * If dir is NULL file will be treated as the whole path.
  * @param handle an alpm handle
@@ -1127,13 +1192,12 @@ time_t _alpm_parsedate(const char *line)
  * @param amode access mode as described in access()
  * @return int value returned by access()
  */
-
 int _alpm_access(alpm_handle_t *handle, const char *dir, const char *file, int amode)
 {
 	size_t len = 0;
 	int ret = 0;
 
-	if (dir) {
+	if(dir) {
 		char *check_path;
 
 		len = strlen(dir) + strlen(file) + 1;
@@ -1148,19 +1212,19 @@ int _alpm_access(alpm_handle_t *handle, const char *dir, const char *file, int a
 	}
 
 	if(ret != 0) {
-		if (amode & R_OK) {
+		if(amode & R_OK) {
 			_alpm_log(handle, ALPM_LOG_DEBUG, "\"%s%s\" is not readable: %s\n",
 					dir, file, strerror(errno));
 		}
-		if (amode & W_OK) {
+		if(amode & W_OK) {
 			_alpm_log(handle, ALPM_LOG_DEBUG, "\"%s%s\" is not writable: %s\n",
 					dir, file, strerror(errno));
 		}
-		if (amode & X_OK) {
+		if(amode & X_OK) {
 			_alpm_log(handle, ALPM_LOG_DEBUG, "\"%s%s\" is not executable: %s\n",
 					dir, file, strerror(errno));
 		}
-		if (amode == F_OK) {
+		if(amode == F_OK) {
 			_alpm_log(handle, ALPM_LOG_DEBUG, "\"%s%s\" does not exist: %s\n",
 					dir, file, strerror(errno));
 		}
@@ -1168,8 +1232,25 @@ int _alpm_access(alpm_handle_t *handle, const char *dir, const char *file, int a
 	return ret;
 }
 
+/** Checks whether a string matches a shell wildcard pattern.
+ * Wrapper around fnmatch.
+ * @param pattern pattern to match aganist
+ * @param string string to check against pattern
+ * @return 0 if string matches pattern, non-zero if they don't match and on
+ * error
+ */
+int _alpm_fnmatch(const void *pattern, const void *string)
+{
+	return fnmatch(pattern, string, 0);
+}
+
 #ifndef HAVE_STRNDUP
 /* A quick and dirty implementation derived from glibc */
+/** Determines the length of a fixed-size string.
+ * @param s string to be measured
+ * @param max maximum number of characters to search for the string end
+ * @return length of s or max, whichever is smaller
+ */
 static size_t strnlen(const char *s, size_t max)
 {
     register const char *p;
@@ -1177,6 +1258,12 @@ static size_t strnlen(const char *s, size_t max)
     return (p - s);
 }
 
+/** Copies a string.
+ * Returned string needs to be freed
+ * @param s string to be copied
+ * @param n maximum number of characters to copy
+ * @return pointer to the new string on success, NULL on error
+ */
 char *strndup(const char *s, size_t n)
 {
   size_t len = strnlen(s, n);