X-Git-Url: http://www.git.cypherpunks.ru/?a=blobdiff_plain;f=doc%2Ffeatures.rst;h=b17e9a8038118ace3e3e456424e29bcc6de2d0ac;hb=7f2e5ef9aef122fb5bdfcb377f0644c7784398a2;hp=76d6830d3913f6872ae1a2f25608f5a9d1916515;hpb=7af6ff46e2927030706fc4c0df0ddeadfa55dd79;p=pyderasn.git

diff --git a/doc/features.rst b/doc/features.rst
index 76d6830..b17e9a8 100644
--- a/doc/features.rst
+++ b/doc/features.rst
@@ -1,7 +1,9 @@
+.. _features:
+
 Features
 ========
 
-* BER/CER/DER decoding, DER encoding
+* BER/CER/DER decoding, strict DER validation, DER/CER encoding
 * Basic ASN.1 data types (X.208): BOOLEAN, INTEGER, BIT STRING, OCTET
   STRING, NULL, OBJECT IDENTIFIER, ENUMERATED, all strings, UTCTime,
   GeneralizedTime, CHOICE, ANY, SEQUENCE (OF), SET (OF)
@@ -10,6 +12,8 @@ Features
   (un)marshall them
 * Python 2.7/3.5/3.6 compatibility
 * Aimed to be complaint with `X.690-201508 <https://www.itu.int/rec/T-REC-X.690-201508-I/en>`__
+* Streaming decoding and encoding capabilities, allowing working with
+  very small memory footprint
 
 Why yet another library? `pyasn1 <http://snmplabs.com/pyasn1/>`__
 had all of this a long time ago. PyDERASN resembles it in many ways. In
@@ -34,6 +38,15 @@ Also there is `asn1crypto <https://github.com/wbond/asn1crypto>`__.
   structures allow BER encoding for the whole message, except for
   ``SignedAttributes`` -- you can easily verify your CMS satisfies that
   requirement
+* Ability to use mmap-ed files, memoryviews, iterators and CER encoder
+  dealing with the writer, giving ability to create huge ASN.1 encoded
+  files without storing all the data in the memory first
+* Ability to decode files in event generation mode, without the need to
+  keep all the data and decoded structures (that takes huge quantity of
+  memory in all known ASN.1 libraries) in the memory
+* ``__slots__``, ``copy.copy()`` friendliness
+* Workability with ``pickle``
+* `Cython <https://cython.org/>`__ compatibility
 * Extensive and comprehensive
   `hypothesis <https://hypothesis.readthedocs.io/en/master/>`__
   driven tests coverage. It also has been fuzzed with
@@ -47,33 +60,7 @@ Also there is `asn1crypto <https://github.com/wbond/asn1crypto>`__.
   automatically set required tags)
 * Descriptive errors, like ``pyderasn.DecodeError: UTCTime
   (tbsCertificate:validity:notAfter:utcTime) (at 328) invalid UTCTime format``
-* ``__slots__``, ``copy.copy()``, ``pickle``, `Cython <https://cython.org/>`__
-  friendliness
-* Could be significantly faster and have lower memory usage
-  For example parsing of CACert.org's CRL (8.48 MiB) on FreeBSD 12.0
-  amd64, Intel Core i5-6200U 2.3 GHz machine, Python 3.5.5/2.7.15:
-
-  .. list-table::
-     :widths: 15 45 20 20
-     :header-rows: 1
-
-     * - Library
-       - Command
-       - Time, sec (Py3/Py2)
-       - Memory used, MiB (Py3/Py2)
-     * - pyasn1 0.4.5
-       - ``der_decode(data, asn1Spec=rfc5280.CertificateList())``
-       - 1257 / 1302
-       - 1327 / 2093
-     * - asn1crypto 0.24.0
-       - ``asn1crypto.crl.CertificateList.load(data).native``
-       - 29.3 / 43.8
-       - 983 / 1677
-     * - pyderasn 4.9
-       - ``CertificateList().decode(data)`` (CertificateList is
-         converted ``pyasn1`` scheme definition)
-       - 27.6 / 32.5
-       - 498 / 488
+* Could be significantly :ref:`faster <performance>` and have lower memory usage
 * :ref:`Pretty printer <pprinting>` and
   :ref:`command-line decoder <cmdline>`, that could
   conveniently replace utilities like either ``dumpasn1`` or