Imported from previous forum
[ original email was from Lakshminarasimhan Rajabather - narasimhan@ssi-technologies.com ]
Is it permitted (or pissible) to use the EncodedText(355)field to send texts with a mix of languages (e.g. Some characters in Japanese and some in English - both using double-byte representation)?
Thanks in advance
The Encoded capability (see Appendix J for more details) was defined to provide a single encoding scheme for a specific message sent to a specific counterparty. In addition to providing an Encoded value, you would provide an English ASCII value in the non-encoded field. You could configure your communication to be custom per counterparty (or message) such that you could send Encoded data in Shift_JIS for one Japanese firm and UTF-8 for Unicode to a French firm.
The MessageEncoding is specified in the Message Header and thus applicable to all of the Encoded fields within that message.
> Is it permitted (or pissible) to use the EncodedText(355)field to send texts with a mix of languages (e.g. Some characters in Japanese and some in English - both using double-byte representation)?
>
> Thanks in advance
>
[ original email was from Lakshminarasimhan Rajabather - narasimhan@ssi-technologies.com ]
Thanks Scott. We are in the process of translating properietary API calls into FIX messages and vice versa. I have a situation where the said API uses one field (binary) to send some text (which could be anything - numbers, english alphabets, japanese characters and/or a mix of all these). Programatically it is difficult to identify the contents of this field and to classify it as Text or Encoded Text.
I was thinking, if I can manipulate the use of EncodedText, by sending Shift_JIS Japanese characters (which contain english alphabets and numerals too) and english alphabets, letting the user be responsible to interpret and display them accordingly (which may not be too difficult). And hence the question. I wasnt sure, however, if this is allowed or not.
I would like to have your views on this. Thanks again for the quick response.
> The Encoded capability (see Appendix J for more details) was defined to provide a single encoding scheme for a specific message sent to a specific counterparty. In addition to providing an Encoded value, you would provide an English ASCII value in the non-encoded field. You could configure your communication to be custom per counterparty (or message) such that you could send Encoded data in Shift_JIS for one Japanese firm and UTF-8 for Unicode to a French firm.
>
> The MessageEncoding is specified in the Message Header and thus applicable to all of the Encoded fields within that message.
>
> > Is it permitted (or pissible) to use the EncodedText(355)field to send texts with a mix of languages (e.g. Some characters in Japanese and some in English - both using double-byte representation)?
> >
> > Thanks in advance
> >
>
You are venturing into some unchartered territory here… Sounds like it could work provided that your counterparties expect and can handle it which would be my primary concern.
If the proprietary API is providing this binary data, maybe you could write some sort of "interpretation filter" which could scan the binary data looking for Japanese characters and assuming Shift_JIS, etc. I, personally, do not have much experience with dealing with character encoding and multi-byte characters.
> Thanks Scott. We are in the process of translating properietary API calls into FIX messages and vice versa. I have a situation where the said API uses one field (binary) to send some text (which could be anything - numbers, english alphabets, japanese characters and/or a mix of all these). Programatically it is difficult to identify the contents of this field and to classify it as Text or Encoded Text.
>
> I was thinking, if I can manipulate the use of EncodedText, by sending Shift_JIS Japanese characters (which contain english alphabets and numerals too) and english alphabets, letting the user be responsible to interpret and display them accordingly (which may not be too difficult). And hence the question. I wasnt sure, however, if this is allowed or not.
>
> I would like to have your views on this. Thanks again for the quick response.
>
>
>
>
> > The Encoded capability (see Appendix J for more details) was defined to provide a single encoding scheme for a specific message sent to a specific counterparty. In addition to providing an Encoded value, you would provide an English ASCII value in the non-encoded field. You could configure your communication to be custom per counterparty (or message) such that you could send Encoded data in Shift_JIS for one Japanese firm and UTF-8 for Unicode to a French firm.
> >
> > The MessageEncoding is specified in the Message Header and thus applicable to all of the Encoded fields within that message.
> >
> > > Is it permitted (or pissible) to use the EncodedText(355)field to send texts with a mix of languages (e.g. Some characters in Japanese and some in English - both using double-byte representation)?
> > >
> > > Thanks in advance
> > >
> >
>