Dan,
Is the content I wrote in the email below sufficient for a doc patch?
An aside: I'm looking at section 8.1 of
<http://www.na-mic.org:8000/websvn/filedetails.php?repname=Dart&path=%2Ftrunk%2FDart.pdf&rev=293>,
and I manually ("eyeballed") compared it with rev 279, and the content
doesn't appear to have changed. I'm wondering if our client tests will
still work. Prashasta and Vignesh, please run some tests with
<http://www.na-mic.org:8000/websvn/filedetails.php?repname=Dart&path=%2Ftrunk%2FResultData%2FTestNewFormat.xml.gz&rev=293>
and let us know what happens.
FWIW, I don't have an Acrobat/pdf editor; maybe freeware ones exist? I
don't know of them.
Also, FWIW: we find it much easier for our projects to track document
changes and history via wikis, for they can much-more-easily reflect
document differences in content (you can diff and display the diffs
graphically) over time. I don't know of a way to do this in .pdf file
controlled in subversion. One could use blame and diffs if it were a text
file, but it's not. So about 4-5 times now, I've looked at
<http://www.na-mic.org:8000/websvn/filedetails.php?repname=Dart&path=%2Ftrunk%2FDart.pdf>
and have wondered: has this file content changed (eg per the manual
exercise above)? When was it changed? What are the differences since the
last changes? What's the history of changes? Who made the changes? How
can I reference specific sections of the document via a web link?
One may argue you can get this from subversion, but when the file is not a
text file, you can't get it for the file's _content_. Even text file
management for these things under subversion pale in comparison with a wiki.
The longer-term solution for these sorts of things, imho: a generic,
XML-based content engine that can be viewed and edited via wikis (or any
other editing mechanism) and can still be auto-generated into a .pdf
file. Then the XML files can be rev-controlled, possibly with subversion,
and then tightly-coupled with a wiki.
Alas, we know of no such thing that does this yet, but our money is on some
DocBook/DocBook-lite thing getting plugged into MediaWiki/TWiki/MoinMoin at
some point down the road.
Also: I obviously can't tell you guys how to do the project; you're making
great stuff and I don't want to get in the way. But I do now how it would
be easiest for myself and my team to contribute *to* the project and learn
how it works.
Having said that, we will get behind whatever method is deemed so by the
project, and this is .pdf-based for now. So we shall fully support .pdf
documentation-content management moving forward as best we can.
-Matt
At 3/7/2006 11:10 AM, Blezek, Daniel J \(GE, Research\) wrote:
Matt,
Rather than put it on the Wiki, can you make corrections to the
Documentation? It is documented in the Dart.pdf file, which we view as
the final repository of all Dart info.
I can incorporate a patch for you.
-dan
-----Original Message-----
From: Matt England [mailto:[EMAIL PROTECTED]
Sent: Tuesday, March 07, 2006 10:17 AM
To: Blezek, Daniel J (GE, Research)
Cc: Matthew England; [email protected]; [EMAIL PROTECTED]; Vignesh
M.P.N.; prashasta gujrati
Subject: RE: [Dart] Looking to create working ctest-to-dart-server link
Dan et al,
At 3/7/2006 07:05 AM, Blezek, Daniel J \(GE, Research\) wrote:
> Sorry this has taken so long to respond. I added an XML example using
> the new Dart format to the SVN repository:
> ResultData/TestNewFormat.xml.gz. It's fairly long, and should include
> almost every type of Result (may not have images).
Ok, great. Is this the file:
http://www.na-mic.org:8000/websvn/filedetails.php?repname=Dart&path=%2Ftrunk%2FResultData%2FTestNewFormat.xml.gz
?
If so, then the format seems to be quite straightforward. I'm attempting
to make some reference docs for our own purpose. If this all works, we'll
wiki-fy it. Can someone verify and/or correct the notes below?
-Matt
The basic "stanza" seems to be the following:
<?xml version="1.0" encoding="utf-8"?>
<!-- Note, we have a submission Jaxor object passed in -->
<DartSubmission version="2.0" createdby="ArchiveTask">
<Site>mysite</Site>
<!-- BuildName will be the concationation of OS and Compiler -->
<BuildName>myplatform-mycompiler</BuildName>
<Track>myfrequency</Track>
<!-- DateTimeStamp is a non-locale specific date/timestamp following
ISO -->
<DateTimeStamp>mydate</DateTimeStamp>
<Test>
<Name>mytestname</Name>
<Status>mybinaryteststatus</Status>
<Measurement name="Completion Status"
type="text/string">mybinarycompletionstatus</Measurement>
<Measurement name="Execution Time"
type="numeric/double">myexectime</Measurement>
<Measurement name="Output" type="text/text">myoutput</Measurement>
</Test>
</DartSubmission>
Where:
mysite = freeform
myfrequency = "Nightly" | ?
mydate = <!-- DateTimeStamp is a non-locale specific date/timestamp
following ISO -->
mytestname = freeform | <.test.name.hierarchy> ?
mybinaryteststatus = "passed" | "failed" | ?
mybinarycompletionstatus = "Completed" | ?
myexectime = floating-point number in seconds
myoutput = freeform | stdout, stderr sections
...although I'm not sure of all the above assertions, and several have
questions marks. (Yes, the above documentation is not formal, and many not
follow XML-DTD description-doc standards...I'm new to things XML in this
area.)
An example instance:
<?xml version="1.0" encoding="utf-8"?>
<!-- Note, we have a submission Jaxor object passed in -->
<DartSubmission version="2.0" createdby="ArchiveTask">
<Site>caleb.crd</Site>
<!-- BuildName will be the concationation of OS and Compiler -->
<BuildName>SunOS-c++</BuildName>
<Track>Nightly</Track>
<!-- DateTimeStamp is a non-locale specific date/timestamp following
ISO -->
<DateTimeStamp>2005-01-03T00:00:00.0-0500</DateTimeStamp>
<Test>
<Name>.Test.Examples.DataRepresentation.Image.Image1Test</Name>
<Status>passed</Status>
<Measurement name="Completion Status"
type="text/string">Completed</Measurement>
<Measurement name="Execution Time"
type="numeric/double">0.641507</Measurement>
<Measurement name="Output" type="text/text">Standard Output:
Standard Error:</Measurement>
</Test>
</DartSubmission>
_______________________________________________
CMake mailing list
[email protected]
http://www.cmake.org/mailman/listinfo/cmake
