diff options
Diffstat (limited to 'portato/eix/parser.pyx')
| -rw-r--r-- | portato/eix/parser.pyx | 78 |
1 files changed, 48 insertions, 30 deletions
diff --git a/portato/eix/parser.pyx b/portato/eix/parser.pyx index 6363b37..a6bcc96 100644 --- a/portato/eix/parser.pyx +++ b/portato/eix/parser.pyx @@ -1,6 +1,6 @@ # -*- coding: utf-8 -*- # -# File: portato/eix/_parser.pyx +# File: portato/eix/parser.pyx # This file is part of the Portato-Project, a graphical portage-frontend. # # Copyright (C) 2006-2010 René 'Necoro' Neumann @@ -20,12 +20,12 @@ For the exact way all the functions work, have a look at the eix format descript __docformat__ = "restructuredtext" cdef extern from "stdio.h": - ctypedef struct FILE: - pass + ctypedef struct FILE int fgetc(FILE* stream) long ftell(FILE* stream) int fseek(FILE* stream, long offset, int whence) + int fread(void* ptr, int size, int n, FILE* stream) int EOF int SEEK_CUR @@ -36,6 +36,11 @@ cdef extern from "Python.h": ctypedef unsigned char UChar ctypedef long long LLong +from python_unicode cimport PyUnicode_DecodeUTF8 +from python_mem cimport PyMem_Malloc, PyMem_Free +from python_exc cimport PyErr_NoMemory +from python_string cimport PyString_FromStringAndSize + from portato.eix.exceptions import EndOfFileException # @@ -54,7 +59,15 @@ cdef int _get_byte (FILE* file) except -1: # Base Types # -cdef LLong _number (object pfile): +cpdef LLong number (object pfile): + """ + Returns a number. + + :param file: The file to read from + :type file: file + :rtype: int + """ + cdef UChar n cdef LLong value cdef int i @@ -84,18 +97,7 @@ cdef LLong _number (object pfile): return value -def number (file): - """ - Returns a number. - - :param file: The file to read from - :type file: file - :rtype: int - """ - - return _number(file) - -def vector (file, get_type, nelems = None): +cpdef object vector (object file, object get_type, object nelems = None): """ Returns a vector of elements. @@ -118,28 +120,44 @@ def vector (file, get_type, nelems = None): cdef LLong i if nelems is None: - n = _number(file) + n = number(file) else: n = nelems return [get_type(file) for i in range(n)] -def string (file): +cpdef object string (object pfile, bint unicode = False): """ Returns a string. - :param file: The file to read from - :type file: file - :rtype: str + :param pfile: The file to read from + :type pfile: file + :param unicode: Return unicode + :type unicode: bool + :rtype: str or unicode """ - nelems = _number(file) + cdef LLong nelems = number(pfile) + cdef FILE* file = PyFile_AsFile(pfile) + cdef char* s + + s = <char*>PyMem_Malloc((nelems + 1) * sizeof(char)) + + if s is NULL: + PyErr_NoMemory() + + try: + if fread(s, sizeof(char), nelems, file) < nelems: + raise EndOfFileException, pfile.name - s = file.read(nelems) + s[nelems] = '\0' - if len(s) != nelems: - raise EndOfFileException, file.name + if unicode: + return PyUnicode_DecodeUTF8(s, nelems, 'replace') + else: # simple string, implicitly copied + return PyString_FromStringAndSize(s, nelems) - return s + finally: + PyMem_Free(s) # # Complex Types @@ -220,8 +238,8 @@ cdef class header: :param file: The file to read from :type file: file """ - self.version = _number(file) - self.ncats = _number(file) + self.version = number(file) + self.ncats = number(file) self.overlays = vector(file, overlay) self.provide = vector(file, string) self.licenses = vector(file, string) @@ -274,12 +292,12 @@ cdef class package: cdef FILE* cfile = PyFile_AsFile(file) cdef long after_offset - self._offset = _number(file) + self._offset = number(file) after_offset = ftell(cfile) self.name = string(file) - self.description = unicode(string(file)) + self.description = string(file, True) # skip the rest, as it is currently unneeded #self.provide = vector(file, number) |