| Performance issues with loading xsds [message #56870] |
Tue, 04 January 2005 15:33  |
Eclipse User |
|
|
|
I am loading some very large XSDs using Eclipse just using the API and am
seeing that the schemas are taking a long time to load. I am calling
resourceSet.getLoadOptions().put(XSDResourceImpl.XSD_TRACK_L OCATION,
Boolean.FALSE) since I don't need to track locations. I was wondering if
there was any other flag I could set that could speed things up
substantially.
The model that I am loading does not need to be manipulated in any way -
it's solely for read purposes only.
Thanks,
Hayden Marchant
|
|
|
|
|
| Re: Performance issues with loading xsds [message #57160 is a reply to message #56925] |
Tue, 18 January 2005 12:19  |
| Eclipse User |
|
|
|
Originally posted by: merks.ca.ibm.com
This is a multi-part message in MIME format.
--------------030506070305010403020103
Content-Type: text/plain; charset=ISO-8859-15; format=flowed
Content-Transfer-Encoding: 7bit
Hayden,
No, there aren't any flags that will significantly boost performance.
But we do hope to have time to do some tuning in the 2.1 development cycle:
82613 <https://bugs.eclipse.org/bugs/show_bug.cgi?id=82613>
Hayden Marchant wrote:
> Elena,
> The XML Parser is not the heavy part of the parsing - it is taking
> about 6% of the total CPU time. I was wondering if there was anything
> else I could do about performance that could really help.
> I'm sure that there must be a tiny little flag that I could set that
> might save me :)
>
> Elena Litani wrote:
>
>> Hi Hayden,
>
>
>> "Hayden Marchant" <hayden.marchant@unicorn.com> wrote in message
>> news:creuj9$uj6$1@www.eclipse.org...
>>
>>> I was wondering if
>>> there was any other flag I could set that could speed things up
>>> substantially.
>>
>
>> If you are parsing your model more than once, you should use the
>> XMLResource.OPTION_USE_PARSER option to specify a parser pool
>> implementation
>> that caches instances of an XML parser for each subsequent parse. The
>> default implemenatation of parser pool is the
>> org.eclipse.emf.ecore.xmi.impl.XMLParserPoolImpl (default constructor is
>> available).
>
>
>>> The model that I am loading does not need to be manipulated in any
>>> way -
>>> it's solely for read purposes only.
>>
>
>> If your application only needs to read and query schemas, you might
>> want to
>> look at the Xerces XML Schema API:
>> http://xml.apache.org/xerces2-j/faq-grammars.html#faq-5
>
>
>> Thanks,
>> Elena
>
>
>
--------------030506070305010403020103
Content-Type: text/html; charset=ISO-8859-15
Content-Transfer-Encoding: 8bit
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-15"
http-equiv="Content-Type">
<title></title>
</head>
<body bgcolor="#ffffff" text="#000000">
Hayden,<br>
<br>
No, there aren't any flags that will significantly boost performance.
|
|
|
| Re: Performance issues with loading xsds [message #593491 is a reply to message #56870] |
Tue, 04 January 2005 16:53 |
| Eclipse User |
|
|
|
Hi Hayden,
"Hayden Marchant" <hayden.marchant@unicorn.com> wrote in message
news:creuj9$uj6$1@www.eclipse.org...
> I was wondering if
> there was any other flag I could set that could speed things up
> substantially.
If you are parsing your model more than once, you should use the
XMLResource.OPTION_USE_PARSER option to specify a parser pool implementation
that caches instances of an XML parser for each subsequent parse. The
default implemenatation of parser pool is the
org.eclipse.emf.ecore.xmi.impl.XMLParserPoolImpl (default constructor is
available).
> The model that I am loading does not need to be manipulated in any way -
> it's solely for read purposes only.
If your application only needs to read and query schemas, you might want to
look at the Xerces XML Schema API:
http://xml.apache.org/xerces2-j/faq-grammars.html#faq-5
Thanks,
Elena
|
|
|
| Re: Performance issues with loading xsds [message #593500 is a reply to message #56898] |
Wed, 05 January 2005 03:04 |
| Eclipse User |
|
|
|
Elena,
The XML Parser is not the heavy part of the parsing - it is taking about
6% of the total CPU time. I was wondering if there was anything else I
could do about performance that could really help.
I'm sure that there must be a tiny little flag that I could set that might
save me :)
Elena Litani wrote:
> Hi Hayden,
> "Hayden Marchant" <hayden.marchant@unicorn.com> wrote in message
> news:creuj9$uj6$1@www.eclipse.org...
>> I was wondering if
>> there was any other flag I could set that could speed things up
>> substantially.
> If you are parsing your model more than once, you should use the
> XMLResource.OPTION_USE_PARSER option to specify a parser pool implementation
> that caches instances of an XML parser for each subsequent parse. The
> default implemenatation of parser pool is the
> org.eclipse.emf.ecore.xmi.impl.XMLParserPoolImpl (default constructor is
> available).
>> The model that I am loading does not need to be manipulated in any way -
>> it's solely for read purposes only.
> If your application only needs to read and query schemas, you might want to
> look at the Xerces XML Schema API:
> http://xml.apache.org/xerces2-j/faq-grammars.html#faq-5
> Thanks,
> Elena
|
|
|
| Re: Performance issues with loading xsds [message #593613 is a reply to message #56925] |
Tue, 18 January 2005 12:19 |
| Eclipse User |
|
|
|
This is a multi-part message in MIME format.
--------------030506070305010403020103
Content-Type: text/plain; charset=ISO-8859-15; format=flowed
Content-Transfer-Encoding: 7bit
Hayden,
No, there aren't any flags that will significantly boost performance.
But we do hope to have time to do some tuning in the 2.1 development cycle:
82613 <https://bugs.eclipse.org/bugs/show_bug.cgi?id=82613>
Hayden Marchant wrote:
> Elena,
> The XML Parser is not the heavy part of the parsing - it is taking
> about 6% of the total CPU time. I was wondering if there was anything
> else I could do about performance that could really help.
> I'm sure that there must be a tiny little flag that I could set that
> might save me :)
>
> Elena Litani wrote:
>
>> Hi Hayden,
>
>
>> "Hayden Marchant" <hayden.marchant@unicorn.com> wrote in message
>> news:creuj9$uj6$1@www.eclipse.org...
>>
>>> I was wondering if
>>> there was any other flag I could set that could speed things up
>>> substantially.
>>
>
>> If you are parsing your model more than once, you should use the
>> XMLResource.OPTION_USE_PARSER option to specify a parser pool
>> implementation
>> that caches instances of an XML parser for each subsequent parse. The
>> default implemenatation of parser pool is the
>> org.eclipse.emf.ecore.xmi.impl.XMLParserPoolImpl (default constructor is
>> available).
>
>
>>> The model that I am loading does not need to be manipulated in any
>>> way -
>>> it's solely for read purposes only.
>>
>
>> If your application only needs to read and query schemas, you might
>> want to
>> look at the Xerces XML Schema API:
>> http://xml.apache.org/xerces2-j/faq-grammars.html#faq-5
>
>
>> Thanks,
>> Elena
>
>
>
--------------030506070305010403020103
Content-Type: text/html; charset=ISO-8859-15
Content-Transfer-Encoding: 8bit
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-15"
http-equiv="Content-Type">
<title></title>
</head>
<body bgcolor="#ffffff" text="#000000">
Hayden,<br>
<br>
No, there aren't any flags that will significantly boost performance.
|
|
|
Powered by
FUDForum. Page generated in 0.07907 seconds
Home