XForms Users Community Group Teleconference -- 09 Sep 2015

<trackbot> Date: 09 September 2015

Orbeon 4.10

https://lists.w3.org/Archives/Public/public-xformsusers/2015Aug/0030

Erik: Not much more to say than is in the email

Call next week [regrets from Steven again]

[No link]

Steven: I'm chairing an ODF meeting next week(in NL), Tues, and Weds
... so shall we skip a week again, or do you want to meet without me?

Philip: I'll be at XML summer school next week, so I won't be able to come anyway.

Steven: So skip a week then.

Editing

https://lists.w3.org/Archives/Public/public-xformsusers/2015Jul/0005.html

Steven: I've been away all week, but this week I'll have more time.

public-xformsusers

[No link]

Steven: Nothing to report, but I'll keep it on the agenda until I hear from Mark.

ACTIONs 2024 - 2029

[See mail list]

<ebruchez> https://lists.w3.org/Archives/Public/public-forms/2015Sep/0000.html

<ebruchez> https://lists.w3.org/Archives/Public/public-forms/2015Sep/0001.html

ACTION-2025 - Deal with the naming over xpath versions

https://lists.w3.org/Archives/Public/public-xformsusers/2015Sep/0000.html

Erik: This was a clarification of a hard-to-understand text

<ebruchez> https://www.w3.org/MarkUp/Forms/wiki/XPath_Expressions_Module#The_XForms_Function_Library

"One question is whether we call "XForms Function Library" only the

functions which are *added* by XForms, or the entire function library,

including the standard XPath (1.0 or 2.0) functions. For now I have assumed

that "XForms Function Library" only contains what is added by XForms.

Steven: I agree with the text as you now have it, that the XForms Function Library is only what is added

ACTION-2029 - Check that integer promotes successfully to double

https://lists.w3.org/Archives/Public/public-xformsusers/2015Sep/0001.html

Erik: We can keep the double parameters
... but returning an empty string for NaN is not backwards compatible.

Steven: But returning an xs:date isn't wither.

Erik: In some cases it wouldn't work, definitely
... The easiest way is to return a string
... but it would look weird in XPath2

Steven: What does XPath2 offer to do things like this?

Erik: Nothing like this.
... We have days-to-date and days-from-date which work together.

Philip: These calculations are difficult. XPath2 calculations are better

<ebruchez> days-to-date(days-from-date(local-date()) + 31)

Steven: XPath2 has days-from-duration, that's how you would do it, right?

<nvdbleek> +1

Steven: I'm inclined to leave our functions as are, returning strings, and let people use XPath2 otherwise

Philip: I agree

<alain> +1

<scribe> ACTION: Erik to make return types of days-to-date and seconds-to-dateTime string [recorded in http://www.w3.org/2015/09/09-forms-minutes.html#action01]

<trackbot> Created ACTION-2030 - Make return types of days-to-date and seconds-to-datetime string [on Erik Bruchez - due 2015-09-16].

trackbot, close action-2029

<trackbot> Closed action-2029.

ACTION-1990: Write text for new instance namespaces processing

http://lists.w3.org/Archives/Public/public-forms/2015Jan/0005.html

https://lists.w3.org/Archives/Public/public-xformsusers/2015Aug/0010.html

<ebruchez> https://lists.w3.org/Archives/Public/public-xformsusers/2015Aug/0012.html

Erik: I replied to Nick's comment
... see the above link
... we want to avoid polluting instances with unwanted namespaces
... especially if you use QNames in attributes
... these won't show up, so you have to include the namespace declaration on the root of the instance

Steven: I think it is worth us saying something explicitly about that.

Nick: We are changing behaviour, but I can live with that.

<ebruchez> https://www.w3.org/MarkUp/Forms/wiki/XForms_2.0#The_instance_Element

Nick: XSLT does something different.
... We have tools to remove redundant namespace declararions.

Steven: We also need an example where there is a default namespace

Erik: And an example where default namespace changes from HTML to XF
... There are drawbacks and advantages to all solutions
... Either way someone's expectation would be broken.

Steven: Does this break anyone's processor?

Nick: I don't think so. I don't think anyone assumes that XF or XH are available in the instance.

Alain: XSLTForms doesn't add those namespaces anyway. I think we might have a prblem with QNames, but I don't think anyone uses that. It is possible to deal with both solutions.

Steven: It wouldn't break your system

Alain: no.

Steven: Nick are you OK with this version?

Nick: I can live with it.

Trackbot, close Action-1990

<trackbot> Closed Action-1990.

ACTION-2020: Propose what to do if serialization in serialize()

fails

https://lists.w3.org/Archives/Public/public-forms/2015Apr/0008

https://lists.w3.org/Archives/Public/public-xformsusers/2015Apr/0006.html

Steven: Why do you prefer an error to an event, Nick?

Nick: I don't think we do that elsewhere.
... and we eventually get an event

Erik: In most cases
... I don't like throwing events while evaluating an XPath function
... it breaks the flow of executions

Steven: So we do strict+error

Erik: I'm not convinced yet.

Steven: Then I suggest we discuss this further next call.

[ADJOURN]

- DRAFT -

XForms Users Community Group Teleconference

09 Sep 2015

Attendees

Contents