- From: <naha@ai.mit.edu>
- Date: Mon, 07 Jan 2002 15:35:36 -0500 (EST)
- To: xsl-list@lists.mulberrytech.com, David Carlisle <davidc@nag.co.uk>
- Cc: www-xml-query-comments@w3.org, Jeni Tennison <jeni@jenitennison.com>
[BTW: I thought we wern't supposed to crosspost] Quoting David Carlisle <davidc@nag.co.uk>: > > > I think that you could argue it both ways > > Yes every argument has two sides: (right and wrong:-) > > > when the function name > > includes the name of a relevant data type from XML Schema, since > those > > data type names use camel case. It depends on whether you think > people > > will be more confused by the naming conventions of functions being > > unpredictable or by not using the same naming convention when > > referring to the built-in data types within function names as you do > > when referring to them elsewhere. > > The old XSLT 1.1 draft went into great detail about mapping between the > lowercase-hypenated-style to the camelCaseStyle in order to preserve th > enature of xpath names. > > If you are mapping xpath hyphenated names to a language that doesn't > allow - in names and conventionally uses camel case, then it is natural > to drop the hyphens and to camel case the component words. In > particular > you _have_ to drop the hyphens when going in that direction. Thus it is > natural when going in the other direction _from_ a camel case naming > convention to do the opposite, to lowercase and add hyphens. Of course > schema names are qnames so could have hyphens, but don't, but I believe > the same principle should apply. I remember when hyphens were replaced by underscores when mapping names to languages which were too empoverished to allow hyphens in names.
Received on Monday, 7 January 2002 15:36:09 UTC