1. 01 Dec, 2023 1 commit
  2. 30 Nov, 2023 2 commits
  3. 29 Nov, 2023 2 commits
  4. 28 Nov, 2023 3 commits
  5. 14 Nov, 2023 1 commit
  6. 13 Nov, 2023 1 commit
  7. 12 Nov, 2023 1 commit
  8. 09 Nov, 2023 3 commits
  9. 10 Oct, 2023 2 commits
  10. 06 Oct, 2023 1 commit
  11. 07 Sep, 2023 1 commit
  12. 01 Sep, 2023 1 commit
  13. 31 Aug, 2023 2 commits
  14. 28 Aug, 2023 1 commit
  15. 24 Jul, 2023 1 commit
  16. 11 Jul, 2023 1 commit
  17. 05 Jul, 2023 3 commits
  18. 15 Jun, 2023 5 commits
  19. 05 May, 2023 2 commits
  20. 04 May, 2023 2 commits
    • Pau Espin Pedrol's avatar
    • tangzhikun's avatar
      fix issue #126: decoding extensible CHOICE error · fc23d52b
      tangzhikun authored
          The constraints ct was not effective for extensible CHOICE, so if ct && ct->flags & APC_EXTENSIBLE was true,
          ct was set to 0. Using normal CHOICE constraint in decoding extensible CHOICE was weird.
          We should change the normal CHOICE constraint to extensible CHOICE constraint when decoding extensible CHOICE,
          the extensible CHOICE constraint was lb=0,ub=specs->tag2el_count-specs->ext_start-1. So the conditional structure
          should be if specs && specs->tag2el_count > specs->ext_start, then decoding the CHOICE which range
          was specs->tag2el_count - specs->ext_start, and add specs->ext_start to value in the last.
      fc23d52b
  21. 18 Nov, 2022 1 commit
  22. 04 Nov, 2022 1 commit
    • Mark's avatar
      rename ATS_BASIC_JER to ATS_JER · 0b34c0f7
      Mark authored
      When JER was added it was based on XER, which has BASIC-XER and
      CANONICAL-XER.
      So originally we had BASIC-JER, which isn't a real thing.
      
      This renames ATS_BASIC_JER to ATS_JER and adds JER in the list of
      encodings in the readme.
      0b34c0f7
  23. 01 Nov, 2022 2 commits
    • Bi-Ruei, Chiu's avatar
      Fix incorrect UPER encoding for empty extensible SEQUENCE in 38.331 · f3a9ca5b
      Bi-Ruei, Chiu authored
      For empty extensible SEQUENCE type, e.g.
      
        searchSpaceType              CHOICE {
          common                       SEQUENCE {
            dci-Format0-0-AndFormat1-0     SEQUENCE {
              ...
            }                                                           OPTIONAL,   -- Need R
            dci-Format2-0                SEQUENCE {
              nrofCandidates-SFI           SEQUENCE {
                aggregationLevel1            ENUMERATED {n1, n2}        OPTIONAL,   -- Need R
                aggregationLevel2            ENUMERATED {n1, n2}        OPTIONAL,   -- Need R
                aggregationLevel4            ENUMERATED {n1, n2}        OPTIONAL,   -- Need R
                aggregationLevel8            ENUMERATED {n1, n2}        OPTIONAL,   -- Need R
                aggregationLevel16           ENUMERATED {n1, n2}        OPTIONAL    -- Need R
              },
              ...
            }                                                         OPTIONAL,   -- Need R
            dci-Format2-1                SEQUENCE {
              ...
            }                                                         OPTIONAL,   -- Need R
            dci-Format2-2                SEQUENCE {
              ...
            }                                                         OPTIONAL,   -- Need R
            dci-Format2-3                SEQUENCE {
              dummy1                       ENUMERATED {sl1, sl2, sl4, sl5, sl8, sl10, sl16, sl20}    OPTIONAL,   -- Cond Setup
              dummy2                       ENUMERATED {n1, n2},
              ...
            }                                                         OPTIONAL    -- Need R
          },
          ue-Specific                    SEQUENCE {
            dci-Formats                    ENUMERATED {formats0-0-And-1-0, formats0-1-And-1-1},
            ...
          }
        }
      
      According to X.691 :
      
        If the sequence type has an extension marker in the "ComponentTypeLists"
        or in the "SequenceType" productions, then a single bit shall first be added to
        the field-list in a bit-field of length one.
      
      So for dci-Format0-0-AndFormat1-0, dci-Format2-1 and dci-Format2-2,
      even they are empty without other fields, there shall be still one extension bit
      to be added.
      
      Use compute_extensions_start() to determine position of extension in
      asn1c_lang_C_type_SEQUENCE_def() to fix this problem.
      f3a9ca5b
    • Bi-Ruei, Chiu's avatar
      Fix code generated from ASN.1 of F1AP v15.4.0 failed to be compiled · 6af78bbd
      Bi-Ruei, Chiu authored
      ProtocolExtensionField.c generated from the following excerpt from ASN.1 of F1AP v15.4.0 :
      
      RRC-Version ::= SEQUENCE {
          latest-RRC-Version BIT STRING (SIZE(3)),
          iE-Extensions      ProtocolExtensionContainer { { RRC-Version-ExtIEs } } OPTIONAL}
      
      RRC-Version-ExtIEs F1AP-PROTOCOL-EXTENSION ::= {
          {ID id-latest-RRC-Version-Enhanced CRITICALITY ignore EXTENSION OCTET STRING (SIZE(3)) PRESENCE optional },
          ...
      }
      
      failed to be compiled because asn_PER_memb_OCTET_STRING_SIZE_3__constr_31 is not generated.
      
      The reason is constraints of types used in information object sets are not processed in asn1fix() stage
      so it's constraint, i.e. (SIZE(3)) in this case, is not generated.
      
      With this fix, OCTET STRING (SIZE(3)) used in ASN.1 of F1AP v15.4.0 can be correctly processed
      and generated code can be compiled.
      
      However, this fix does not work for all sorts of types.
      
      For example, if information object refer to a type like this : SEQUENCE (SIZE(3)) OF INTEGER
      it will be failed to compiled because asn_IOS_RRC_Version_ExtIEs_1_rows[] refers to non-existing
      asn_DEF_SEQUENCE_OF.
      6af78bbd