We must not only consider files in the current directory (ADF.USIM)
but also in its sub-directories. This requires us to be able to
determine the path we need to traverse between the currently selected
file (EF.UST) and the respective file in some other directory,
which is implemented via CardFile.build_select_path_to().
Change-Id: I61797fefa9dafa36a8a62c11aa2cfaeecb015740
We had service annotations only for ADF.USIM so far, but not for
the related sub-directories.
Change-Id: Iaa56a26ba53eaf18fce14845ae07a27c52a2c58a
Note: The code doesn't make use of them in any reasonable way yet!
The existing code had the following serious problems:
* when trying to update EF.SST or EF.IST, it would write to EF.UST !
* shell commands were called ust_* even for the EST/IST files
Let's introduce the proper separation between what is shared and what
is file-specific.
Change-Id: Ie55669ca37a4762fac9f71b1db528ca67056e8dd
This command performs a consistency check between the services activated
in EF.UST/EF.IST and the files that should (or should not) be
active/selectable for the given service.
Produces output like:
Checking service No 48 (inactive)
ERROR: File EF(EF.MWIS) is selectable but should not!
Checking service No 49 (active)
ERROR: File EF(EF.CFIS) is not selectable (SW=6a82) but should!
Change-Id: Iea7166959e2015eb8fa34d86036560c9e42ce4d3
This allows us [in a future patch] to perform consistency checking,
whether files exist for services not activated in EF.{UST,IST} or
vice-versa: Services are activated by files are not present or
deactivated.
Change-Id: I94bd1c3f9e977767553000077dd003423ed6dbd1
This can be populated by card profiles with the SST/IST/UST service
that is associated with the file.
Change-Id: I3b3f74b691368fa09967ecb377a9f7a6d8af7869
pytlv is a nightmare of shortcomings, let's abandon it in favor of
our own meanwhile-created pySim.tlv. This has the added benefit
that unknown tags finally no longer raise exceptions.
Change-Id: Ic8e0e0ddf915949670d620630d4ceb02a9116471
Closes: OS#5414
This happens e.g. when selecting the ARA-M applet on sysmoISIM-SJA2:
pySIM-shell (MF)> select ADF.ARA-M
-> 00a4040409 a00000015141434c00
<- 9000:
Traceback (most recent call last):
File "/space/home/laforge/.local/lib/python3.9/site-packages/cmd2/cmd2.py", line 2064, in onecmd_plus_hooks
stop = self.onecmd(statement, add_to_history=add_to_history)
File "/space/home/laforge/.local/lib/python3.9/site-packages/cmd2/cmd2.py", line 2494, in onecmd
stop = func(statement)
File "/space/home/laforge/projects/git/pysim/./pySim-shell.py", line 750, in do_select
fcp_dec = self._cmd.rs.select(path, self._cmd)
File "/space/home/laforge/projects/git/pysim/pySim/filesystem.py", line 1314, in select
select_resp = f.decode_select_response(data)
File "/space/home/laforge/projects/git/pysim/pySim/filesystem.py", line 193, in decode_select_response
return self.parent.decode_select_response(data_hex)
File "/space/home/laforge/projects/git/pysim/pySim/filesystem.py", line 378, in decode_select_response
return profile.decode_select_response(data_hex)
File "/space/home/laforge/projects/git/pysim/pySim/ts_102_221.py", line 796, in decode_select_response
t.from_tlv(h2b(resp_hex))
File "/space/home/laforge/projects/git/pysim/pySim/tlv.py", line 231, in from_tlv
(rawtag, remainder) = self.__class__._parse_tag_raw(do)
File "/space/home/laforge/projects/git/pysim/pySim/tlv.py", line 258, in _parse_tag_raw
return bertlv_parse_tag_raw(do)
File "/space/home/laforge/projects/git/pysim/pySim/utils.py", line 208, in bertlv_parse_tag_raw
if binary[0] == 0xff and len(binary) == 1 or binary[0] == 0xff and binary[1] == 0xff:
IndexError: bytearray index out of range
EXCEPTION of type 'IndexError' occurred with message: 'bytearray index out of range'
Change-Id: I910e6deba27d1483dff1e986c89f1a1b2165f49b
We have a number of integers with variable-length encoding, so
add a Construct for this. Naming inspired by GreedyBytes.
Related to https://github.com/construct/construct/issues/962
Change-Id: Ic6049b74ea3705fda24855f34b4a1d5f2c9327f7
Our hand-written JSON so far is using snake_case identifiers,
while the JSON generated by the pySim.tlv classes use the class
names as keys, which LooksQuiteDifferent.
So let's auto-convert the CamelCase into something that reflects
our existing notion.
Change-Id: Id55929ef03dc48cb668e6ba7e99b6b291680a42f
We had a mixture of tab and 4space based indenting, which is a bad
idea. 4space is the standard in python, so convert all our code to
that. The result unfortuantely still shoed even more inconsistencies,
so I've decided to run autopep8 on the entire code base.
Change-Id: I4a4b1b444a2f43fab05fc5d2c8a7dd6ddecb5f07
EF.DIR can not only contain the AID + Label of TS 102 221, but can
also contain any of the DOs specified in ISO7816-4. Let's imoplement
this based on the modern pySim.tlv parser
Change-Id: I875eb49e1f0370428c2eae69af84f5483bd5b1fc
Closes: OS#5410
As we've seen in recent patches, this has been a source of bugs, so
let's be tolerant and deal with both.
Change-Id: I0a5ec2a860104ffe4524c647105a42505ac394d6
In Change-Id I6d7c1bf49a8eaf3d8e50fb12888bf3d5b46b6c55 we fixed the
filesystem code to assume the self._tlv memper is a reference to a
class, and not an instance (as this is what the majority of the code
did).
However, it seems thre wer two instances where we actually had _tlv
reference an instance. Change that to class so it's the same all over
the code base.
Change-Id: Ie4878ad6a92feafe47e375c4f5f3f198921e1e95
So far, we only returned an array of service numbers like
[ 2, 4, 5, 9 ] which is not very friendly to the human reader.
In EF.SST we already had more verbose decoding including a description
of each service. Let's add the same principle to EF.UST, EST and IST
The same output above now looks like this:
{
"1": {
"description": "Local Phone Book",
"activated": false
},
"2": {
"description": "Fixed Dialling Numbers (FDN)",
"activated": true
},
"3": {
"description": "Extension 2",
"activated": false
},
"4": {
"description": "Service Dialling Numbers (SDN)",
"activated": true
},
"5": {
"description": "Extension3",
"activated": true
},
"6": {
"description": "Barred Dialling Numbers (BDN)",
"activated": false
},
"7": {
"description": "Extension4",
"activated": false
},
"9": {
"description": "Incoming Call Information (ICI and ICT)",
"activated": true
}
}
Change-Id: I34f64d1043698dc385619b2fdda23cb541675f76
At least on Debian 10 and unstable, I'm getting this error for pylint:
************* Module pySim.utils
pySim/utils.py:570:1: E0611: No name 'strxor' in module 'Crypto.Util.strxor' (no-name-in-module)
despite it clearly existing:
>>> import Crypto.Util.strxor
>>> Crypto.Util.strxor.strxor
<built-in function strxor>
So let's suppress the related pylint error.
Change-Id: Iea89e758782a569be953d19892028f083a92c2f1
Add file specific command `update_imsi_plmn` to EF_IMSI to replace
the mcc and mnc part of the imsi for use in bulk_script(s)
Change-Id: I9662ff074acf9dc974ae4c78edac44db286e98fc
when encoding the AcT value bit 11 is correctly set
when NG-RAN is present in the string representation,
however the decoding of bit 11 was missing.
Adds tests for the decoder as well.
Change-Id: I910df28c4c59ec94cce9603377786325f6d8c1a3
When probing applications on a card by running select_adf_by_aid()
SwMatchError exceptions indicating the non exsistance of that
application on the card should be ignored.
Change-Id: I3aa7deaf46bdf352a201c3089b3714405a06f281
When printing applications found by probing for a specific AID, then the
wrong variable is used to print the AID.
Change-Id: I3d5ec28e46fe00c0d793a1d9ef0a0e0900649a4d
A profile can cover lots of different applications. Those applications
may not exist on all card models. To exclude applications that are not
installed on the particular card EF.DIR is evaluated. However, there may
be applications that are not registered in EF.DIR but supported by the
profile. To cover those as well, lets try to select the applications we
do not see in EF.DIR. If selecting works we know that the application
exists on the card and we can include them in the RuntimeState.
Change-Id: I3fa77a68664fe50d690a18adfb1ae1a88a189827
There is no need for us to expand a partial AID to the full AID before
selecting that ADF. The UICC specifications permit AID selection by
prefix only. So we could pass the prefix to the card, and the card would
do the prefix matching. In order to avoid problems with cards that fail
to do the prefix matching themselves we will still do the AID
completion, but in case we cannot complete the AID (AID not listed in
EF.DIR), we will try with the AID prefix anyway.
From the API user point of view, this allows us to select applications
not listed in EF.DIR
Change-Id: I0747b4e46ca7e30bd96d76053765080367ac1317
The method decode_select_response does not access any property of the
object. This means the method can be static.
Change-Id: Idd7aaebcf1ab0099cd40a88b8938604e84d8a88b
At the moment the non checking send_apdu() method is used when records
are read. Lets use read_record_checksw so that we get an exception in
case there is a problem to read the specified record.
Change-Id: I9fc411e1b12e8d9fd89b9964209808c0706011bd
The method decode_select_response just calls the function
_decode_select_response. But the function _decode_select_response
is not called from any other location, so we can move it into the
profile class.
Change-Id: Icf0143f64ca7d1c1ebf60ba06585f7afc1ac0d11
UICC and old SIM cards can be difficult to tell apart without prior
knowledge of the card. The ATR won't tell if the card is UICC or not.
The only remaining option is to try out if the card is able to handle
UICC APDUs. The same is true for 2G SIM cards. It is not guranteed that
every UICC card will have 2G functionality.
Lets add functionality to match a profile to the currently plugged card
by actively probing it.
Lets also add another profile to distinguish between UICC-only cards and
UICC cards that include SIM functionality.
Change-Id: If090d32551145f75c644657b90085a3ef5bfa691
Related: OS#5274
The function select_adf_by_aid first searches for the complete AID in
the set of AIDs that were read from EF.DIR. Lets put this task into a
separate helper method
Change-Id: I88447d47bc96d0d4ff5cea694b46e854232cdf86
This introduces support for talking to the ARA-M application on a card,
as specified in the GlobalPlatform "Secure Element Access Control"
specification v1.1.
Change-Id: Ia9107a4629c3d68320f32bbd4dd26e1f430717da
The table that holds the status word descriptions is initialized as an
empty list '[]'. This is not correct since the interpret_sw method
processes this data as dictionary, so lets initialize the sw member with
an empty dict '{}' when not status word description is given.
Change-Id: I3cae83f0f6ab274546991ecd14425f094b2816b2
Related: OS#5274
The class byte and the select control bytes are different for SIM cards
and UICC cards. Lets define those parameters in the card profile, so
that we always get the correct parameters depending on which profile we
use.
Change-Id: I2d175e28bd748a4871b1373273b3a9be9ae8c4d0
Related: OS#5274
When the runtime state is created there is already some interaction with
the card. Lets make sure that the card is in a defined state when we
leave the constructor of the RuntimeState.
Change-Id: I986204964903069bcce781afdbf3c5d26682b749
Related: OS#5274
When the profile does not define any ADFs, then do not try to read any
AIDs. This is the case for old non UICC SIMs for example.
Change-Id: I8cfbee1d23e9f99461fa5f4fbf92c1a0929c50bf
Related: OS#5274
There are some problems with the usage of decode_select_response. At the
moment the ADF files overload the related method to provide decoding of
the select responses as per 3gpp TS 102 221. However, this also means
that the decoder is only available under ADF.USIM and ADF.ISIM. DF.GSM
and DF.TELECOM also overload the decoder method, just like an ADF would
do. This decoding method is then implemented as per 3gpp TS 51 011.
Since this a a problem on UICCs, the method detects the magic byte 0x62
that can be found at the beginning on every select response of an UICC
to defer to the TS 102 221 decoding method. TS 51 011 defines the first
two bytes of the select response as RFU. This at least problematic.
To solve this there should be a default method for
decode_select_response in the profile, which can be used if no file
overloads it with a specific decoder. ADFs use specific decoders, but
everything else should use the default decoder. When we deal with an
UICC, we expect the select response to be consistantly conform to TS
102 221, if we deal with a clasic sim we expect responses as per TS 51
011 only.
Since it is still possible to replace the select response decoder we
still have the opportunity to have custom select response in cartain
DFs and ADFs should we need them.
Change-Id: I95e33ec1755727dc9bbbc6016ce2d99a9e66f214
Related: OS#5274
The computed length of the file may be negative, when the offset exceeds
the file length. When this is the case, return none
Change-Id: I2c017c620254fae188022851ef3b670730aab503
The transport layer provides a method send_apdu_checksw to send APDUs
and to be sure the SW is the expected one. Given that, there is no need
to verify the SW manually. The exception of send_apdu_checksw will catch
the problem and also display the SW in a human readable form.
Change-Id: I9ce556ac0b7bb21c5c5a27170c32af0152255b79
Related: OS#5275
CardProfileSIM is currenty instantiated directly. However, it should be
implemented as class and then instaniated later like CardProfileUICC
Change-Id: I37d49b11a07ce5a80d1a703fab4620b7d1ecb25b
Currently we call the reset_card and get_atr methods directly at the
transport layer via the private _scc and _tp object of the card. This is
a violation. Fix and use the reset methods that are already in the
SimCard object.
Change-Id: I0e9d2a62a42a7387e7ca69d2ae830782a61aed89
There are situations where it is necessary to modify the class byte and
the selection control bytes of a card at runtime. This should not be
done by accessing the properties of the _scc object directly. The
modification of those properties should be done via a set method
instead.
Change-Id: Ifd8aa2660e44a18d28945d070419612eff443e78
We cannot simply skip anything that has 0xFF as first byte to detect
the padding after the end of a TLV object:
0xFF may very well be a valid first octet of a multi-octet TAG:
Tags of private class (11) with constructed (1) payload will have 0xFF
as first octet.
So let's expand the check to only detect padding in case of either only
a single byte FF being left, or two FF following each other [with
whatever suffix].
Change-Id: I5d64ce9ef1d973804daabae0b15c2e2349e6fab9
When calling from_dict() on a hierarchy of nested BER_TLV_IE,
only the first/outer layer of TLV_IE_Collection would get its
'decoded' initialized correctly from the dict. Subsequent layers
were not, as the 'decoded=' was passed as parameter during instance
initialization. If we first instantiate the class and then call the
from_dict() method, the recursive initialization down the full hierarchy
works as expected.
Change-Id: I795a33ed8dfa8454dc9079c189ab7b2ba64a3b72
There are instances where a TLV IE is used as just a flag, i.e.
length zero and no value part. In those situations, it would require
a lot of boilerplate code to require the TLV_IE class definitions to
have _to_bytes/_from_bytes methods that do nothing.
So instead, add a shortcut: If we want to encode 'None', then return
b'', and if we want to decode b'' return None.
Change-Id: Ie8eb2830e8eefa81e94b8b8b157062c085aeb777
GSM-R SIM cards have an additional directory (DF.EIRENE) with a number
of files. This is all specified in the following document:
UIC Reference P38 T 9001 5.0 "FFFIS for GSM-R SIM Cards"
Change-Id: I4034d09292a08d277d4abcbed9a0ec2808daaacb
In change Id410489841bb9020ddbf74de9114d808b1d5adb6, the RuntimeState
class automatically adds additional files to the CardApplications for
ISIM and USIM. This works only once. The second time an exception will
be thrown because the added files are already in the CardApplication.
Currently there is no way generate new card applications during
initialization because the card applications are just objects that are
created once in ts_31_10x.py. Lets turn them into classes and create the
objects during initialization. This way we get fresh objects when we
re-initialize.
Change-Id: Ibb4f6242e7a92af84a905daa727b1b87016e7819
Similar to the fix in Ie1aeaab29701946233ed73db3331039690d695da
for update_binary(), read_binary() also contained a bug when treating
non-zero offsets.
Change-Id: Ic5c2f0ad1c1ec9c4e9c97e72895382f7b6fa9470
Related: OS#5254
Currently a card must be present in the reader until the user can enter
pySim-shell. Removing and plugging another card is in theory already
possible, but then the new card will operate on the old card and runtime
state object. It might also be useful to enter pySim-shell before the
card is plugged to execute some other commands for preperation before.
So lets allow to "equip" pySim-shell with a card and rs object at
runtime.
Related: SYS#5617
Change-Id: I9cf532d9da8203065463c7201e7064de6c7ab1b5
The select response decoder is using b2h() wrongly. b2h expects
a bytearray but we call it with an integer. In the following two
lines we try to convert an integer to an integer.
Change-Id: Ib6448d3bd7a0dc7f25e5ee82a42266b3313e2a95
In Icc240d5c8c04198640eb118565ea99f10ba27466 we introduced support for
writing files > 255 bytes by splitting the write into multiple chunks.
However, at the same time, that commit broke support for writing data at
non-zero offsets. Unfortunately, this is used extensively within
pySim-prog e.g. for writing K + OP/OPc data to sysmoISIM-SJA2 and sysmoUSIM-SJS1
cards.
This commit fixes the related problem.
Change-Id: Ie1aeaab29701946233ed73db3331039690d695da
Fixes: Icc240d5c8c04198640eb118565ea99f10ba27466
Closes: OS#5254
Before:
EXCEPTION of type 'NotImplementedError' occurred with message: ''
After:
EXCEPTION of type 'NotImplementedError' occurred with message: 'EF(EF.SST) encoder not yet implemented. Patches welcome.'
Change-Id: Ie8a10a8847f7c7c6a3332fb9f78de18c9f7f41d0
The card_detect function in cards.py allows to specify the card type or
use the hints "auto" and "auto_once" to trigger autodetection of the
card. However, "auto_once" has no effect and is not used by any caller,
so lets remove it.
Change-Id: Iea726f51e5ddb43d8a4da2672552fff38e29b006
* introduce type annotations
* introduce + derive implementations from base class
* move shared code to base class
Change-Id: I7168506cbebb1ebb67f47453419b860824912051
The FairwavesSIM programming fails when the card is accessed with USIM
APDUs. To keep it working temporarly switch to SIM APDUs during
programming.
Change-Id: I8f02625d2b620ecdf4b2afc27a8750119b707152
The constructor gets an sl object on initalization. The card handler
will then carry out the reader operation wait_for_card().
In cases where an mechanically automated card reader is used it may
be useful to go without those operations and let the caller carry out
the appropriate reader operations. So Lets make the sl object
optional for the CardHandlerAuto class. If it is not present, simply
do not carry out the pre programmed reader operation.
Change-Id: I0f793aec51751b7c7b87d55b66326cce9970274e
Related: SYS#5617
Make sure that a reader is disconnected before connecting it. This will
efectively prevent resource leakage in the lower PCSC layers when the
reader is connected multiple times during bulk provisioning
Change-Id: I266e56f2330da25c680a76f4c0ca630a38e1f61b
There may be corner cases where an execption contains no error message.
In this case it might still be helpful to display the type of the
exeption calss to get at least an idea of what kind of error we are
dealing with.
Change-Id: I6e6b3acd17e40934050b9b088960a2f851120b26
In OOP, we usually use capital letters for class names. The card handler
class should be no execption.
Change-Id: I4b2c06b1c607c993c9aaf0d57ad2352bb6b36e74
The T0 protocol (selected in transport/pcsc.py) does not support extended APDU, so 255 bytes is the maximum number of bytes that can be transmitted at a time. We can divide large data into 255 byte chunks. The read_binary function already has code to read more than 255 bytes, so we can just adapt it to the update_binary function.
Change-Id: Icc240d5c8c04198640eb118565ea99f10ba27466
There are the classes IsimCard and UsimCard, which inheret from Card,
which is the base class for a normal non ISIM/USIM simcard. Card also
has methods in it that are related to simcards, so it is not just any
"Card", it is a SimCard and should be called that way.
Change-Id: I2077ded44bc2297b8d478c5bd1895951b494efcc
Harald Welte
Bjoern Riemer
Steve Markgraf
Philipp Maier
Vadim Yanitskiy