...

...

...

...

"CD" == Charles Duffy <charles@dyfis.net> writes:

CD> The newline option just indicates which kind of newlines to use in CD> places where newlines would normally be selected -- it doesn't make CD> lists automatically newline-separated. CD> Maybe you want something more like this: CD> declare option output:format "no"; CD> declare option output:method "text"; CD> string-join(("one", "two", "three"), " ") What an embarrassment that one must hardcode record separators or whatever. I feel like I am back forty years ago with http://en.wikipedia.org/wiki/NOS_(software) Let me out of here! It should be as easy as $ seq 1 3 1 2 3 $ perl -le 'print for 1..3' 1 2 3 Please fix this!

Dear talk@x-query.com, there should be a way to get 'items' back out of Xquery, one per line, without having to hardwire 0x0A (you know, $ perl -wle 'printf "%08b\n" , ord"\n"' 00001010 into the code, as discussed in https://mailman.uni-konstanz.de/pipermail/basex-talk/2012-March/002738.html ) ...or require a post-processor.

...

...

...

...

"DL" == David Lee <dlee@calldei.com> writes: DL> There is, but you have to call XQuery natively instead of via "the command line"

Running the output through a tag stripper, perl -pwle 's/<[^>]+>//g' sounds a 1000 times easier than trying to figure out how to do it whatever is that way you mention.

...

...

...

...

"LREQ" == Liam R E Quin <liam@w3.org> writes: LREQ> On Sat, 2012-03-10 at 22:56 +0000, David Lee wrote: There is, but you have to call XQuery natively instead of via "the command line" LREQ> ??!?

LREQ> you can use string-join( your query here, " ") Haven't had to make my own newlines since using those Control Data computers back in the 80's. But if you insist... LREQ> or you may be able to use serialization options from the command line. That would be nice.

On Sun, 2012-03-11 at 10:14 +0800, jidanni@jidanni.org wrote:

Haven't had to make my own newlines since using those Control Data computers back in the 80's. But if you insist...

XQuery is not really a line-oriented language. I don't agree with you that "it should be as easy as seq 1 3" any more than it should be easy to process XML documents recursively an element at a time with sed, or to use grep only within certain elements. Unix is line-oriented, XQuery is XML-oriented. This is true regardless of whether you used Control Data computers. Having said that, I can imagine a BaseX feature request - or maybe even an XQuery feature request - for how to separate items in a sequence when serializing. Such a request could be filed against the 3.0 serialization spec at https://www.w3.org/Bugs/Public/ and the XQuery and XSLT Working Groups will discuss it. Liam -- Liam Quin - XML Activity Lead, W3C, http://www.w3.org/People/Quin/ Pictures from old books: http://fromoldbooks.org/

It looks like the CLP of DB2 gives you the behaviour you are asking for http://publib.boulder.ibm.com/infocenter/db2luw/v9r5/index.jsp?topic=%2Fcom.... But as mentioned before this is outside XQuery itself it's the environment , API , or processor that adds this bit I personally feel it would have been a better choice to have used new lines ( whatever they are ) instead of spaces as the default XDM serialization separator for sequences but that's really a tiny thing in the grand scheme of things. Xmlsh does this but it's vastly more useful to NOT serialize sequences but rather keep them in internal form so You can pick them apart Before handing them off to a non XDM aware process. Hence the suggestion of using an API that can get at the results of XQuery aka XDM before serialization. Xmlsh also supports this .. Getting the results of any XDM process in an internal form without serialization. But that's again the province of the processor and/or API. Not XQuery itself. Again back to the unanswered questions What processor are you using What results do you expect and why What is your end goal Maybe there is a better way to accomplish your end goals then text serialization tweaks to XDM Sent from my iPad (excuse the terseness) David A Lee dlee@calldei.com On Mar 10, 2012, at 10:24 PM, "jidanni@jidanni.org" <jidanni@jidanni.org> wrote:

OK, I submitted https://www.w3.org/Bugs/Public/show_bug.cgi?id=16311 However as I have a rather hard time expressing myself, perhaps those who know what I am saying could add some detail to the bug. In the rare case the detail is wrong, I will chime in there. Thanks. _______________________________________________ talk@x-query.com http://x-query.com/mailman/listinfo/talk

OK, I submitted https://www.w3.org/Bugs/Public/show_bug.cgi?id=16311 However as I have a rather hard time expressing myself, perhaps those who know what I am saying could add some detail to the bug. In the rare case the detail is wrong, I will chime in there. Thanks.

I've added the following suggestion to the Bugzilla entry: I've looked up the XQuery Serialization Spec.: Step 3 in Section 2, "Sequence Normalization", contains the sub-sentence "separated by a single space" [1], which would probably have to be extended to take an additional serialization parameter into account. An XQuery 3.0 expression that uses the parameter could e.g. look as follows: declare option output:separator " "; (1 to 10) Opinions are welcome. Christian [1] http://www.w3.org/TR/xslt-xquery-serialization-30/#serdm

On Sat, 2012-03-10 at 22:56 +0000, David Lee wrote:

There is, but you have to call XQuery natively instead of via "the command line" ??!?

My reading was the OP wanted newlines without having to add them in XQuery code. That is beyond XQuery itself so my suggestion of using a native API. But alas I belive I've misunderstood the entire post. -David dlee@calldei.com

All I know is using $ cat file.xq (: Make bus timetable going westbound past Zaokeng :) declare option db:parser "html"; declare option output:method "html"; (: etc., not poorly supported "text" though :) declare function local:d($M){ doc(concat("http://www.fybus.com.tw/data/", $M)) }; let $line:=("1/067b.htm","city/90.htm","city/207.htm", "city/259.htm", "city/267.htm") let $name:=("6572"," 90"," 207", " 259", " 267") let $correction:=(23,4,23,2,29) for $l at $c in $line, $i in distinct-values(local:d($l)//*:tr/*:td[position()=2 and contains(.,":")]) let $t := xs:time(concat($i, ':00')) + xs:dayTimeDuration('PT1M') * $correction[$c] order by $t return <TD>{if(contains($l,"259") or (contains($l,"90")and contains($i,":00")))then "平" else "每",substring(string($t), 1, 5), $name[$c]}</TD> I need to convert to text mode by hand: $ basex file.xq|perl -pwle 's/<[^>]+>//g' As all text serialization in Xquery can't deal with the same line concept... unless one hardwires the line separators (0x0A) in via string-join, which doesn't sound much like serialization to me.

...

...

...

...

"AW" == Andrew Welch <andrew.j.welch@gmail.com> writes: AW> So the easy way, is to just return a single item using string-join. Which means we text wanters don't get to enjoy the benefits of all that serialization work one single bit.

AW> ...which suggests you just want indent set? All I know is I just added an example of what I want "a flip of a switch" declaration change to cause in https://www.w3.org/Bugs/Public/show_bug.cgi?id=16311#c2