Recent changes to feature-requests

#39 Add ability to have content in the prolog

Stanimir Stamenkov — Sat, 15 Apr 2023 06:44:19 -0000

[22]    prolog     ::=      XMLDecl? Misc* (doctypedecl Misc*)?
[27]    Misc       ::=      Comment | PI | S

What type of content, and where more specifically you expect it to be allowed?

In any case, SAX is just an API specification. The actual behavior is governed by:

The XML specification; and
The parser/processor implementation in use.

#40 Error message Element ... cannot have character [children], because the type's content type is element-only.

Stanimir Stamenkov — Sat, 15 Apr 2023 06:31:14 -0000

SAX is just an API specification. The exact message is emitted by the parser implementation in use:

at org.apache.xerces.util.ErrorHandlerWrapper.createSAXParseException(Unknown Source)

You should bring this up with the Xerces project.

Error message Element ... cannot have character [children], because the type's content type is element-only.

Bruno Dirkx — Thu, 06 Sep 2018 11:30:28 -0000

Hello
I received this error:
Caused by: org.xml.sax.SAXParseException; cvc-complex-type.2.3: Element 'formula' cannot have character [children], because the type's content type is element-only.
at org.apache.xerces.util.ErrorHandlerWrapper.createSAXParseException(Unknown Source)

The error being the use of illegal characters between tags in the XML, in this case a pipe symbol.
Example: [...] <formula><attribute name="abc" value="123"/>|<attribute name="def" value="456"></formula>

The error message was not very clear, our feature request is to have a more clear error message.

Add ability to have content in the prolog

calsosta — Sat, 29 Aug 2015 03:46:59 -0000

Content should be allowed in the prolog.

Sample

Is any feature exist in SAX to ignore case of xml tags??

Anonymous — Tue, 23 Jun 2009 05:37:26 -0000

Hi,

I have a requirement, where in while parsing my xml data validating the xml with XSD, validation is failing when tags are not matched with the case. So, I am facing problem because of this. Please let me know if there is any feature available to ignore case while validating.

Attributes should retain line number information

Anonymous — Tue, 05 May 2009 22:25:44 -0000

For the given XML fragment:

The org.xml.sax.Attributes API does not allow line number information to be queried for individual attributes.

Could you please consider an org.xml.sax.ext.Attributes3 interface and introduce something like the following?

public int getLineNumber(int index);

request for an inbuilt xml chunker

cbiswa — Sun, 15 May 2005 03:56:31 -0000

I am sure a lot of people use the SAX parse for parsing
really large xml documents. In most cases, these large
documents consist of repeating chunks that can be
processed one at a time.

If there was an inbuilt xml chunker provided along with
the helper classes, most applications would really
benefit from it instead of each having to reinvent the
wheel.

I am attaching example XmlChunker code that I have
found very useful for most xml parsing needs. It is a
simple framework consisting of 2 classes -
"XmlElement", "XmlChunker" and one interface -
"XmlHandler".

I think they could be used as a start for something that
should really be a part of the helper classes.

Thanks.

SAXParser truncating value of some records

Arun — Sun, 13 Feb 2005 07:46:15 -0000

Im parsing a long XML file at runtime in my web
application. Im facing a problem here that after parsing
a long set of data towards the end the SAX Parser is
truncating the value of an entry. Its a long XML file
generated dynamically and there are around 83 records
in the resultset. The parser parses fine till the 81st
record and at 82nd entry value it truncates the value.
The sample xml is follows:

<XMLRESULTSET>
<ROW>
<COL name="ERRORCODE"></COL>
<COL name="ERRORDESCRIPTION"></COL>
<COL name="COMPANYNO">3345672</COL>
<COL name="NEWJOBPROVIDER">0</COL>
<COL name="NOOFJOBRECORDS">83</COL>
<MULTIVALUEDCOL name="RECORDS">
<ENTRYSET>
<ENTRY name="HIGHLIGHT">0</ENTRY>
<ENTRY name="JOBOPENINGRECORDNO">3124</ENTRY>
<ENTRY name="JOBTITLE">Admin Manager</ENTRY>
<ENTRY name="JOBOPENINGSTATUS">1</ENTRY>
<ENTRY name="JOBOPENINGPOSTDATE">2004-12-
14</ENTRY>
<ENTRY name="NOOFAPPLICANTS">0</ENTRY>
<ENTRY name="NOOFCANDIDATES">0</ENTRY>
<ENTRY name="NOOFINTERVIEWEES">0</ENTRY>
</ENTRYSET>
.
.
.
(82nd)
<ENTRYSET>
<ENTRY name="HIGHLIGHT">0</ENTRY>
<ENTRY name="JOBOPENINGRECORDNO">3140</ENTRY>
<ENTRY name="JOBTITLE">Execut
Representative./wireline</ENTRY>
<ENTRY name="JOBOPENINGSTATUS">1</ENTRY>
<ENTRY name="JOBOPENINGPOSTDATE">2004-12-
22</ENTRY>
<ENTRY name="NOOFAPPLICANTS">0</ENTRY>
<ENTRY name="NOOFCANDIDATES">0</ENTRY>
<ENTRY name="NOOFINTERVIEWEES">0</ENTRY>
</ENTRYSET>
<ENTRYSET>
<ENTRY name="HIGHLIGHT">0</ENTRY>
<ENTRY name="JOBOPENINGRECORDNO">3141</ENTRY>
<ENTRY name="JOBTITLE"> Execut
Representative./wireline</ENTRY>
<ENTRY name="JOBOPENINGSTATUS">1</ENTRY>
<ENTRY name="JOBOPENINGPOSTDATE">2004-12-
22</ENTRY>
<ENTRY name="NOOFAPPLICANTS">0</ENTRY>
<ENTRY name="NOOFCANDIDATES">0</ENTRY>
<ENTRY name="NOOFINTERVIEWEES">0</ENTRY>
</ENTRYSET>
</MULTIVALUEDCOL> </ROW>
</XMLRESULTSET>

Now when my java application parses the above xml, it
provides truncated result for 82nd record date as 2004-
12 not 2004-12-22 without giving any exception. Also its
parsing the next record correctly that is 83rd date
comes fine as 2004-12-22. I have tried all possible
debugging and procedure to get this sorted out, but
nothing result. Also, Im reading all entry value in
StringBuffer as I read in other forum threads that its
preferred than normal String object. Following is my
block of code with method: characters().

StringBuffer sbInhalt = null;
String entryVal = ;

public void startElement(String uri, String localName,
String qName, Attributes attrs) throws
SAXParseException {
sbInhalt = new StringBuffer();
}

public void characters(char[] ch, int start, int length) {

if(name.equalsIgnoreCase("COL")){
colValue = new String(ch,start,length).trim();

}else if(name.equalsIgnoreCase("ENTRY")){
sbInhalt = sbInhalt.append(ch, start, length);
entryValue = sbInhalt.toString();

System.out.println("In Char Start,
Length==>>"+start+", "+length);
System.out.println("EntryVal in
Char==>>"+entryValue);
}
}

public void endElement(String uri, String localName,
String qName)
throws SAXParseException{

System.out.println(The result Entry in Buff
Format==>>+ sbInhalt); //This is Printing truncated
82nd Entry
System.out.println(The result
Entry==>>+entryVal); //This is Printing truncated 82nd
Entry
entryVal = ;
sbInhalt = null;
}

Also, I observed that after about every 16384 start
character position, the start value gets resets. And in
this 82nd record the date value returns length as 7 not
10. Is these any way I can control the char[] length
passed in SAXParser? Is there any configuration for SAX
in this effect. Im completely puzzled with this behavior
of my parser application. Any suggestions and solution
will be highly appreciated.

SAXParseException does not implement Locator

Lars Marius Garshol — Mon, 28 Jul 2003 10:15:45 -0000

SAXParseException does not implement Locator, which means that
when I want to do the following

try {
parser.parse(inpsrc);
} catch (SAXParseException e) {
// do something useful with info, and record location it occurred in
}

I have the problem in the catch block that I can't create a new
SAXParseException and keep the Locator info easily, because
although the exception wants a Locator as the second constructor
arg, the SAXParseException is not a Locator. Also, if I want to keep
the information around somewhere else I can't use the copy
constructor of LocatorImpl, because, again, SAXParseException is not
a Locator.

The only solution is to do

Locator loc = new LocatorImpl();
loc.setPublicId(e.getPublicId());
loc.setSystemId(e.getSystemId());
// and so on...

which is not exactly pleasant.

parsing a portion of a single xml document at a time

Anonymous — Sun, 08 Jun 2003 20:21:08 -0000

I'd like to see a :

void parse( String chunkOfXmlDoc, boolean isFinished );

along the same lines as Expat or libxml's primary parse
routines. I'd like the parser to parse what it can of
this chunk and perform whatever callbacks it is able
to, then return. Then I want to be able to call the
same parser some time in the future with another string
which will be appended to whatever is remaining of the
parser's buffer and parsing will continue where it left
off.

I hope this doesn't sound unreasonable. I'm able to
simulate this behavior with multiple threads or by
hacking my xml parser of choice, but it would be nice
if there was a standard extension simiiar to this.