The choice member is called df-5gprose but the header is called
'df-5g-prose-header' (note the '-' between '5g' and 'prose'). WTF.
Change-Id: I86004ac2e18a187c26c5e470344908512d21fb9e
Sometimes the struct member is called like df-telecom, but in other
cases it's called df-df-saip with a double 'df' in front. That makes
no sense, but we have to deal with it from our constructors...
Change-Id: If5e670441f03a47fa34e97a326909b24927c12f7
We're creating a 'pyosmocom' pypi module which contains a number of core
Osmocom libraries / interfaces that are not specific to SIM card stuff
contained here.
The main modules moved in this initial step are pySim.tlv, pySim.utils
and pySim.construct. utils is split, not all of the contents is
unrelated to SIM Cards. The other two are moved completely.
Change-Id: I4b63e45bcb0c9ba2424dacf85e0222aee735f411
This is the only way we can make sure pylint has all required
information about imports from packages we depend upon.
Change-Id: I29582aa3d7f9ace9ce832d5b907420aaf14881fb
************* Module pySim.esim.es2p
pySim/esim/es2p.py:107:19: E1101: Class 'datetime' has no 'toisoformat' member (no-member)
Change-Id: Ib762792d595048bf6d7d6f5acbe2715f137ae5bb
With this change, the ProfileElementSequence object will maintain a
representation of the filesystem hierarchy of the eSIM profile. Every
file that is added by a ProfileElement will add a FsNode into that tree,
and each FsNode will point to the File object for the respective file.
This allows us to find files by their path, as well as add files by
path.
Change-Id: I2caadc24b1087855f23f3c57cdf8dabbf81757c0
The self.files member is a dict. Hence we should use those dict
keys when [re]building the decoded dict. The previous code ignored
it and re-constructed the key from File.pe_name - but that's not
always identical.
Change-Id: I0e6c97721fb1cfc6b5c21595d85bd374d485b573
Store a back-reference to the PE-Sequence in the PE object; this is
neccessary for some upcoming patches, e.g. to determine the position in
the sequence, access the global filesystem hierarchy, etc.
Change-Id: I24b692e47e4dd0afb5a17b04d5e0251dded3d611
Having AKA specific code in the generic ProfileElement base class dated
back to when we didn't have a ProfileElementAKA subclass.
Change-Id: Icd332183758b8ef20a77507b728f5e455698def0
When populating a File from a FileTemplate, let's make sure we
* correctly treat the maximum file size for BER-TLV files
* respect the default value pattern / repeat pattern
* respect the high_update flag.
Change-Id: I3ba092e0893f53a18264dff5fa37b12ccd9bd47e
So far the prefix_match() required a string argument; let's also
permit another OID object to be passed; we internally convert that
to string.
Change-Id: I0feb7782d1813cc46ec78f170eb0fce804aebe3a
we so far supported construction of the Path object from a string or
a list of strings. Let's also add the option of constructing it from a
path consisting of a list of integer FID values.
Change-Id: Ia7e9375b3258d1fbfdc892cefba3e3bbe841c550
before this change, structure == 'ber_tlv' was missing the
file_type == working_ef attribute. So for linear_fixed, transparent
and cyclic, the file_type attribute was present, but for ber_tlv it was
missing. This is illogical from a user point of vie and makes downstream code
potentially more complex, as it cannot match on working_ef for all EF
types.
Change-Id: If0076cc6dd35a818c08309885f6ef1c1704052c6
The SAIP specification is very weird in a way that it treats the DF and
EF descriptions as some kind of flat structure without describing the
hierarchy. So when creating a DF, sometimes it should be created below
the current DF, and sometimes it should be adjacent next to the current
DF.
Let's introduce
* a 'ppath' property of FileTemplate to indicate if a file is anything
but a direct sibling of the 'base DF' of the PE
* an 'extends' property of ProfileTemplate to indicate that a given
template does not have its own 'base DF', but that its contents merely
extends that of another ProfileTemplate
* a 'parent' property of ProfileTemplate to indicate a parent
ProfileTemplate below whose 'base DF' our files should be placed.
Change-Id: Ieab4835cd21008b289713784c0eb7170af2ccfb9
The Card.file_exists() method is only called by legacy pySim-{read,prog}
when it wants to determine if it can read/write a file. Therefore
it actually doesn't only want to know if the file exists, but also
if it's not deactivated.
Change-Id: I73bd1ab3780e475c96a10cd5dbdd45b829c67335
Closes: OS#6530
This method can be used to expand the default value pattern of the
file system template for the file to the specified (record, file) length.
Change-Id: Id3eb16910c0bdfa572294e14ca1cd44ca95ca69f
The default value must contain '...' to indicate a variable-length
default value section, not '..'
Change-Id: I8d78278065c145b86460acf8eb723babe777c4f6
The SAIP specification version implicitly determines which filesystem
templates (or versions thereof) are supported. So if a given eUICC
states it implements SAIP version 2.3.0, then we have to translate
this into which template versions that means. The new SaipSpecVersion
and its derived classes do exactly that.
Change-Id: I3a894c72c22e42bd2067e067be80a67197ad1bf2
The SAIP data format is inherently flat and doesn't intrinsically
have an idea of the tree-like structure of a filesystem. However,
if we want to (for example) convert a physical USIM into an eSIM
profile, we need to find the template for a given file, where the file
is identified by its path.
Let's expose a path property of the FileTemplate object, and populate
that when creating the FileTemplate as part of a ProfileTemplate.
Change-Id: Ie145ba159081daf8fbfa544f6d4248f05b7eea96
We previously only permitted this for transparent files (TR), but
file size can of course also be specified for BER-TLV files.
Change-Id: Ie007cf2ccde0a17d0fb853a96b833f064ae52c59
So far we only had the EF.ICCID and EF.PL within our UICC card profile.
However, a classic GSM SIM card is not an UICC, so the CardProfileSIM
also needs those files.
To avoid circular dependencies, move the definitions from ts_102_221.py
to ts_51_011.py
Change-Id: I6eaa5b579f02c7d75f443ee2b2cc8ae0ba13f2fe
Closes: OS#6485
The existing logic is wrong. How we call from_dict() doesn't differ if
a member IE itself contains a nested collection: We always must pass a
single-entry dict with the snak-case name of the class to from_dict().
Change-Id: Ic1f9db45db75b887227c2e20785198814cbab0f5
Fixes: OS#6453
Historically, to_dict and from_dict were not symmetric; this has been
fixed in I07e4feb3800b420d8be7aae8911f828f1da9dab8 in December 2023.
This however broke the ara_m legacy use of the from_dict() methods.
We've just introduced a from_val_dict() method in
I81654ea54aed9e598943f41a26a57dcc3a7f10c2, let's make use of it.
Change-Id: I3aaec40eb665d6254be7b103444c04ff48aac36d
We put those in ts_102_221 because that's where ProprietaryInformation
is defined, and we don't want to risk circular dependencies.
Change-Id: I526acfeacee9e4f7118f280b3549fd04fdb74336
This fixes a long-standing bug in the FileDescriptor IE class which so
far only supported decoding, but not encoding of BER-TLV file
descriptors.
Change-Id: I598b0e1709ee004bcf01a53beb91f68470e1f3da
This command exports the entire filesystem state as one JSON document,
which can be useful for storing it in a noSQL database, or for doing a
structured diff between different such dumps.
It's similar to "export", but then reasonably different to rectify a
separate command.
Change-Id: Ib179f57bc04d394efe11003ba191dca6098192d3
There are some situations where we want to work with a type-name-wrapped
dict that includes the type information, and others where we don't want
that. The main reason is that nested IEs can only be reconstructed if
we can determine the type/class of the nested IE from the dict data.
Let's explicitly offer {to,from}_val_dict() methods that work with
the value-part only
Related: OS#6453
Change-Id: I81654ea54aed9e598943f41a26a57dcc3a7f10c2
In the read and write command implementations, we used to catch
lower-layer exceptions (usually SwMatchError) and "translate" that into
a value error, only to add more information to the exception. This
meant that higher-layer code could no longer detect this was actually
a SwMatchError exception type.
Let's instead use the add_note() method to amend the existing exception,
rather than raising a new one of different type.
Change-Id: Ic94d0fe60a8a5e15aade56ec418192ecf31ac5e7
The new methods allow programmatic resolution of nested IEs from
a parent, assuming there's only one child of a given type (which is
often but not always the case).
Change-Id: Ic95b74437647ae8d4bf3cdc481832afb622e3cf0
When loading a ProfileElement from its DER-ecoded format, populate
a dict with a pySim.esim.saip.File object for each file.
Change-Id: Ie2791c10289eb28daed2904467b0c5e5b11c94c2
Until Change-Id Ifba1048e3000829d54769b0420f5134e2f9b04e1 the TAR
output was working for implicit tar. With said commit we fixed it
for explicit tar but broke implicit tar.
With this commit it works for both implicit and explicit TAR.
Change-Id: I76133b0e02996a138257f3fba5ceb0d2fc6fad80
In the past, we always wrapped a HexAdapter around bytes-like data in
order to make sure it's printed as hex-digits. However, now that we are
doing JSON output it's much easier to let the pySim.utils.JsonEncoder
take care of this in a generic way.
We should do a similar migration all over pySim (pySim-shell,
filesystem, etc.) - but for now only do it in the low-hanging fruit of
pySim-trace aka pySim.apdu
Change-Id: I0cde40b2db08b4db9c10c1ece9ca6fdd42aa9154
Let's factor out the "automatic processing using _tlv / _construct" as a
separate method. This way we enable a derived class to first call that
automatic processing method, and then amend its output in a second step.
Change-Id: I1f066c0f1502020c88d99026c25bf2e283c3b4f5
This is requird to make some definitions available to USIM / ts_31_102
without introducing circular dependencies.
Change-Id: I32e29f400d2da047e821bf732316b21805b5a1e2
As SGP.22 states, the handleNotification endpoint uses HTTP status 204,
not 200 (due to its empty body).
Change-Id: I890bdbd3e1c4578d2d5f0367958fdce26e338cac
This implements the first parts of the "GlobalPlatform Remote
Application Management over HTTP Card Specification v2.3 - Amendment B,
Versoin 1.2". Specifically, this patch covers the TLV definitions for
the OTA message used for HTTPS session triggering.
This also adds some more unit test coverage to pySim.cat, based on
real-world data that was captured nested inside the HTTPS Administration
session triggering parameters.
Change-Id: Ia7d7bd6df41bdf1249011bad9a9a38b7669edc54
this way, the eUICC will send us notifications whenever our profiles are
enabled/disabled/deleted.
Change-Id: I2861290864522b691b30b079c7c2e1466904df2d
"other" notifications (enable, disable, delete) contain ECDSA
signatures that also need verification.
Change-Id: If610058b7af6f9fc7822576c93f9970e2ce9aba9
The ES9+ interface is not only used for downloading eSIM profiles, but
it is also used to report back the installation result as well as
profile management operations like enable/disable/delete.
Change-Id: Iefba7fa0471b34eae30700ed43531a515af0eb93
SGP.22 is quite clear in that handleNotification shall return an empty
HTTP response body. Let's make sure we comply to that and don't report
a JSON response.
Change-Id: I1cad539accbc3e7222bfd4780955b3b1ff694c5b
For example, the ES9+ handleNotification function is defined with an
empty response body, so we cannot unconditionally assume that every HTTP
response will contain a JSON "header" value.
Change-Id: Ia3c5703b746c1eba91f85f8545f849a3f2d56e0b
TS 102 221 specifies that (in case of a class 4 command) and as SW
62xx or 63xx, we should send a GET RESPONSE just like in the 61xx
case in order to get the respective response.
As we don't really know if it's a case1/2/3/4 command in the
pySim.transport, let's always send the GET RESPONSE in case SW 62xx or
63xx are received. It shouldn't hurt - in the worst case there's no
response available...
Change-Id: Ibb1398194a16fc1f1f9bc46af6c66fb6575240cd
SMS cannot exceed 140 bytes, and TS 31.115 explicitly states that larger
messages must use multi-part SMS, which we don't yet implement here.
Change-Id: I8a1543838be2add1c3cfdf7155676cf2b9827e6e
while it's true that in situations where response_status == 'por_ok'
we are guaranteed to have a 'secured_data' key in the dict, its value
could well be b'', which in turn causes us to run into an exception,
calling a decoder on an empty byte value; let's avoid that.
Change-Id: I7c919f9987585d3b42347c54bd3082a54b8c2a0a
the COMPACT-TLV variant is a TLV variant that ISO7816 uses for encoding
tag and length into a single octet. This is used (for example) in ATR
historical bytes.
Let's add support for this to our pySim TLV encoder/decoder.
Change-Id: I9e98d150b97317ae0c6be2366bdaaeaeddf8031c
The ProfileHeader PE contain lists of template-oids and services that
are mandatory in this profile. Let's add methods that can be used to
(re-) compute those lists based on the actual PE contents of the
sequence.
The idea is that during programmatic construction of a profile, those methods
would be called after appending all PEs, just before encoding the
profile as DER.
Change-Id: Ib43db8695c6eb63965756364fda7546d82df0beb
Let's avoid the copy+paste in the subclass constructors and initialize the profile
element header in the base class constructor.
Change-Id: I6e69ae1f0d33d963247fc506db33b3840c10c19a
There are e.g. templates for usim and for opt-usim, and they should not
be confused with each other. Let's reflect that in the naming.
Change-Id: Ic6d04ce3172dc969c6b8c018b8d305eb6fd3f550
Harald Welte