I think I have come across a concrete example of this. It's from a standard called vCard 3.0 (RFC 2426 - http://tools.ietf.org/html/rfc2426). A backslash is used to a) escape itself; b) escape in-scope delimiters; c) indicate an embedded linefeed. The backslash would need removing for a) and b) but not for c).

ESCAPED-CHAR = "\\" / "\;" / "\," / "\n" / "\N")
; \\ encodes \, \n or \N encodes newline ; \; encodes ;, \, encodes ,
Regards

Steve Hanson
Architect, IBM DFDL
Co-Chair, OGF DFDL Working Group
IBM SWG, Hursley, UK
smh@uk.ibm.com
tel:+44-1962-815848

From: Steve Hanson/UK/IBM
To: "Cranford, Jonathan W." <jcranford@mitre.org>,
Cc: "dfdl-wg@ogf.org" <dfdl-wg@ogf.org>, dfdl-wg-bounces@ogf.org, Mike Beckerle <mbeckerle.dfdl@gmail.com>
Date: 04/06/2014 08:53
Subject: Re: [DFDL-WG] data where inactive escape character is to be retained

WG call 3rd June: The generalised use case is perhaps speculative, so it was agreed not to change the DFDL spec to handle this unless a concrete use case emerges.

Regards

Steve Hanson
Architect, IBM DFDL
Co-Chair, OGF DFDL Working Group
IBM SWG, Hursley, UK
smh@uk.ibm.com
tel:+44-1962-815848

From: "Cranford, Jonathan W." <jcranford@mitre.org>
To: Mike Beckerle <mbeckerle.dfdl@gmail.com>, "dfdl-wg@ogf.org" <dfdl-wg@ogf.org>,
Date: 20/05/2014 18:11
Subject: Re: [DFDL-WG] data where inactive escape character is to be retained
Sent by: dfdl-wg-bounces@ogf.org

All,

I’ll chime in with an observation and few more details on how Roger Costello got around a similar problem.

Observation
An escape block allows the escapeBlockEnd to be escaped with the escapeEscapeCharacter, while allowing the escapeEscapeCharacter itself to appear in the data without any special semantics as long as it is NOT followed by escapeBlockEnd. (From section 13.2.1 of the spec: “On parsing the dfdl:escapeBlockStart is removed from the beginning of the data and dfdl:escapeBlockEnd is removed from end of the data and any dfdl:escapeEscapeCharacters are removed when they precede a dfdl:escapeBlockEnd.”)

This is really similar to the problem that Mike posed to the group below, where an escape character is sometimes an escape character but sometimes isn’t.

While an escape block might not be suitable in all circumstances, the original problem that sparked Mike’s post was amenable to using an escape block, and that is how Roger Costello got around the problem.

Some more details
Roger Costello was using a quotation mark (“) as the initiator and terminator for quoted values in a data format. In this format, quotation marks can be escaped with a backslash (\); however, within a quoted string, the data could have a backslash as a normal data character (e.g. \n, representing two characters, not a single newline character).

Roger posed his challenge to the Daffodil team, and then Mike created the example below to demonstrate the problem to the DFDL WG. In contrast to Mike’s example, Roger was having the problem with initiators and terminators, not a separator. At the time, we thought that that an escape block couldn’t be applied to the data format in question, so Mike may have altered the problem in order to prevent an escape block from clouding the issue as posed to the WG.

It turns out, after more analysis, that an escape block could be used, and that solved the problem:
escapeBlockStart=”"” escapeBlockEnd=”"” escapeEscapeCharacter=”\”

Closing Observations
In general, DFDL supports two different escape schemes with different behavior for the escape character.
* When escapeKind=”escapeCharacter”, the escape character is always an escape character.
* When escapeKind=”escapeBlock”, the escape character (escapeEscapeCharacter) is only an escape character in front of escapeBlockEnd.

In this case, we were able to use an escape block to model the data format. While there may be a data format that has a character that is sometimes an escape character and sometimes isn’t, without a real world example, I echo Mike’s hesitance to add this feature to DFDL.

HTH,

Jonathan Cranford

From: dfdl-wg-bounces@ogf.org [mailto:dfdl-wg-bounces@ogf.org] On Behalf Of Mike Beckerle
Sent: Friday, May 02, 2014 2:47 PM
To: dfdl-wg@ogf.org
Subject: [DFDL-WG] data where inactive escape character is to be retained

We have data that has ; as separator and has fields that look like this:

abcd \; efgh

That's a single field.
The backslash escapes the ; so that the data is abcd ; efgh.
This same data set also has

abcd \n efgh
Here the backslash precedes an ordinary non-delimiter. The data is supposed to be abcd \n efgh. That is, this data set requires the backslash to be retained in the data when it is not preceding the start of a delimiter.
Am I missing something or is it impossible to model this?

It would seem there needs to be a flag to indicate whether the escape characters that don't actually escape a delimiter are to be retained or not.

Mike Beckerle | OGF DFDL Workgroup Co-Chair | Tresys Technology | www.tresys.com
Please note: Contributions to the DFDL Workgroup's email discussions are subject to the OGF Intellectual Property Policy
-- dfdl-wg mailing list dfdl-wg@ogf.org https://www.ogf.org/mailman/listinfo/dfdl-wg

Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number 741598.
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU

Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number 741598.
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU