This Annex gives an overview on how the national language extension mechanism of the GSM 7 bit default alphabet works. This Annex shows how a message with an indication of the Turkish National Language Identifier is decoded, but the same principles apply to other languages.

This example outlines the behaviour of both supporting and non-supporting receiving entities where the Turkish National Language Single Shift Table is indicated in the received message. In this example there is no locking shift mechanism used in parallel. A non-supporting receiving entity will ignore the National Language Single Shift IE, and decode the message contents using the GSM 7 bit default alphabet Table defined in subclause 6.2.1, including possible escape characters to the GSM 7 bit default alphabet extension Table specified in subclause 6.2.1.1. For example the Turkish word "Türkçe" will be displayed as "Türkce". A receiving entity that supports the Turkish National Language Single Shift Table will detect a National Language Single Shift IE in a TP User Data Header. This IE tells the receiving entity that the single shift mechanism is used. A supporting receiving entity will notice the language code, in this example coded as '0000 0001', and therefore use the Turkish National Language Single Shift Table defined in subclause A.2.1 instead of the GSM 7 bit default alphabet extension Table defined in subclause 6.2.1. If the next character is any character except <escape>, then the GSM 7 bit default alphabet Table is used for the decode. If the next character is <escape> then the Turkish language specific Table is used for the decode of the one character that follows the <escape>. This process will be repeated until the end of the received message, or until the end of the current segment of a concatenated message. The Language selection at the start of a message takes 4 octets which correspond to five 7 bit characters which reduces the maximum number of characters per single message to 155. Thereafter, the number of characters within that single message will be dependent upon the number of times a character is used that is within the National Language Single Shift Table. Every character used from the National Language Single Shift Table will need an additional character to identify the escape to the National Language Single Shift Table. The available 155 character capacity of a single message will therefore be reduced accordingly. This reduction of overall message length also applies when using characters from the GSM 7 bit default alphabet extension Table (see subclause 6.2.1.1) when the National Language Single Shift IE is not used.

This example outlines the behaviour of both supporting and non-supporting receiving entities where the Turkish National Language Locking Shift Table is indicated in the received message. A non-supporting receiving entity will ignore the National Language Locking Shift IE, and decode the message contents using the GSM 7 bit default alphabet defined in subclause 6.2.1, including possible escape characters to the GSM 7 bit default alphabet extensions specified in subclause 6.2.1.1. A receiving entity that supports the scheme will detect a National Language Locking Shift IE in a TP User Data Header. This IE tells the receiving entity that the locking shift mechanism is used. If no National Language Single Shift IE is indicated additionally to the National Language Locking Shift IE, then the whole message is decoded using the National Language Locking Shift Table defined for Turkish language in Table 6.2.1.2.4.1. If, in addition to the National Language Locking Shift IE (which may be for Turkish or another language), a National Language Single Shift IE for the Turkish language is indicated, then <escape> makes an exception to the use of the National Language Locking Shift Table for the Turkish or another language. In that case a character following <escape> is decoded using the National Language Single Shift Table for the Turkish language, after which the use of the National Language Locking Shift Table for the Turkish or another language is resumed until the next <escape> or the end of the message is met. The Language selection at the start of a message takes 4 octets which corresponds to five 7 bit characters which reduces the maximum number of characters per single message to 155, unless the National Language Single Shift IE has also been included, in which case there is a further reduction of 3 octets making 7 octets in total, which corresponds to eight 7 bit characters, which reduces the maximum number of characters per single message to 152. Thereafter, if the single shift mechanism is used additionally to the locking shift mechanism, the number of characters within that single message will be dependent upon the number of times a character is used that is within the National Language Single Shift Table. Every character in the National Language Single Shift Table will use an additional character. The available 152 character single message length will therefore be reduced accordingly. This reduction of overall message length also applies when using characters from the GSM 7 bit default alphabet extension Table (see subclause 6.2.1.1) when the National Language Single Shift IE is not used.