|
|
These test results are presented as three separate sets of data. The small
document tests use four collections of small documents:
- soaps (.4-1.4 KB) - SOAP request and response messages, taken
from the Apache Axis interoperability test results. These use namespaces
extensively, some attributes along with character data content.
- fms (each about 5 KB) - RDF files giving new release information
from freshmeat.net. Heavy use of namespaces, a few attributes along with
character data content.
- ants (.5-9.9 KB) - Ant build.xml files taken from Jakarta
Taglibs projects. No namespaces, heavy attributes and comments with little
character data content.
- webs (.2-36 KB) - Jakarta Taglibs taglib.tld and web.xml
files. No namespaces, character data content with no attributes.
The mid-sized document tests use four documents in the 100-200 KB size
range. The documents used were:
- soap.xml (131 KB) - a generated SOAP document containing a
large list of values. Some namespaces and attributes, flat structure consisting
of simple elements with short character data content. This document was generated
by Aleksander Slominski as a SOAP test case.
- much_ado.xml (197 KB) - the Shakespeare play marked up as
XML. No namespaces or attributes, flat structure consisting of simple elements
with relatively long character data content. This came from Jon Bosak's collection
of documents.
- periodic_table.xml (114 KB) - periodic table of the elements
in XML. No namespaces and light attribute usage, fairly complex structure
consisting mainly of elements with short character data content. This document
originated with Elliotte Rusty Harold.
- xml.xml (192 KB) - the XML specification as XHTML, with the
DTD reference removed and all entities substituted (necessary for some of
the models used in the tests). This was chosen as typical of document presentation
markups, with heavy mixed content. No namespaces, light use of attributes.
Note that this file cannot be included in the distribution due to
the licensing, which requires that it be distributed only in unmodified form
- to use this, you'll need to remove the DTD reference and substitute &#xxx
values for entities yourself.
Finally, a single document is used to test performance for large documents:
- weblog.xml (2.9 MB) - web server access log file formatted
as XML. This consists of approximately 10K elements representing page hits,
each containing several child elements with character data content for the
fields of information. No namespaces and no attributes. This document was
received from David Mertz.
|
