So the use case that drives the question is syslogd format. Part of the syntax is a whitespace separated list of pairs like so: foo="stuff with spaces" bar="more stuff with spaces and equal = signs" The spaces separate the pairs, the quotation marks are required, not optional, so they're not escapeBlockStart/End, they're initiator and terminator. There's a sequence with space separator here. Inside that is recurring "pairs" containing name and content separated by "=". Zero or more pairs. Content has an initiator and terminator which are double quotes. The spaces inside the string content are *not* escaped. Nor equal signs. emptyValueDelmiterPolicy is 'both', non-nillable, so nilValueDelimiterPolicy is not relevant. Seems to me a parser for this does not need escaping of the spaces or = that appear inside the content, but the DFDL spec can only express parsing these if those escapes are provided. Am I interpreting the spec correctly in this case? That because the surrounding groups have space and = separators, that the content must escape these if they appear? 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 <http://www.ogf.org/About/abt_policies.php> On Wed, Aug 16, 2017 at 11:28 AM, Steve Hanson <smh@uk.ibm.com> wrote:
In general, enclosing construct's delimiters are also relevant. When scanning for the value of an element with a terminator, there are some circumstances where there might not be a terminator: - nil value delimiter policy says there is no terminator - empty value delimiter policy says there is no terminator - element is optional so if you find enclosing construct delimiter as first character the element is missing
So you *could* design a wholly delimited format where enclosing construct delimiters never needed escaping but it would be a bit restrictive in practice. Formats that I have seen where enclosing construct delimiters are not escaped usually have fixed length fields.
Regards
Steve Hanson
IBM Hybrid Integration, Hursley, UK Architect, *IBM DFDL* <http://www.ibm.com/developerworks/library/se-dfdl/index.html> Co-Chair, *OGF DFDL Working Group* <http://www.ogf.org/dfdl/> *smh@uk.ibm.com* <smh@uk.ibm.com> tel:+44-1962-815848 <+44%201962%20815848> mob:+44-7717-378890 <+44%207717%20378890>
From: Mike Beckerle <mbeckerle.dfdl@gmail.com> To: "dfdl-wg@ogf.org" <dfdl-wg@ogf.org> Date: 16/08/2017 15:48 Subject: [DFDL-WG] clarification question on terminators vs. enclosing group separators/terminators Sent by: "dfdl-wg" <dfdl-wg-bounces@ogf.org> ------------------------------
The DFDL Spec says:
12.3.2 *dfdl:lengthKind 'delimited'*
On parsing, the length of an element with dfdl:lengthKind 'delimited' is determined by scanning the datastream for the delimiter.
The data stream is scanned for any of
· the element's terminator (if specified)
· an enclosing construct's separator or terminator
· the end of an enclosing element designated by its known length
· the end of the data stream
So if an element has a terminator, are the enclosing constructs' separator or terminator also relevant? Or is ONLY the element's own terminator relevant for scanning, and hence, only the element's own terminator must be escaped if it appears in the content.
For example, in a space-separated group, an enclosed element has a terminator ";". When parsing that element, do spaces have to be escaped if they appear in the content, or does only the terminator ";" have to be escaped?
Strictly speaking it seems enclosing delimiters shouldn't have to be escaped, because the data must have the ";", and spaces are only significant as separators after finding the ";" terminator.
Mike Beckerle | OGF DFDL Workgroup Co-Chair | Tresys Technology | *www.tresys.com* <http://www.tresys.com/> Please note: Contributions to the DFDL Workgroup's email discussions are subject to the *OGF Intellectual Property Policy* <http://www.ogf.org/About/abt_policies.php> -- 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