Thanks Boris and all the others. Radical it is.... and thanks for the offer. However, We're quite bound to use Altova set of XML tools because their code-generators create the complete API for our products (OCR/ICR Recognition engines). The API must be available both in C++, Java and C# --- your data binding system (so I believe) is limited to C++ on a small variety of platforms, whereas our integrators spread around many different platforms (Windoze of all kinds, Unix's etc).
The API created by Altova generator is very comprehensive to integrators, and is automatically synchronized with my schema changes. Moreover, its Java, C# and C++ versions are almost identical semantically, which is, again, very good for our integrators. Last, and not least --- there is a generator for the documentation... Only problem is that their C++ /xerces version was not designed carefully, and has this major drawback, that in our case created an intolerable memory bloating. I am deep into rewriting their mid-layer of code, which communicates the generated API with xerces. I must say, xerces is not particularly helpful, or documented, or provides reasonable sample code... I Thanks god (and all of you) for this mailing list! Motti Shneor Senior Software Engineer Orbograph Ltd. [EMAIL PROTECTED] http://www.orbograph.com -----Original Message----- From: news [mailto:[EMAIL PROTECTED] On Behalf Of Boris Kolpackov Sent: Wednesday, December 13, 2006 9:55 AM To: [email protected] Subject: Re: DOMDocument memory bloating problem Hi Motti, "Motti Shneor" <[EMAIL PROTECTED]> writes: > My alternative is to completely rewrite the code-generator to use > independent documents for each object, which is a lot of work. > > Any ideas? One (radical) idea is to use a better XML data binding tool, for example CodeSynthesis XSD (full disclosure: I am involved in the development). While XSD also uses Xerces-C++ as an underlying XML parser, it keeps DOMDocument around only for a short period during parsing and serialization. Also each new document parsing or serialization action uses new DOMDocument instance so there is no memory bloat problem. http://www.codesynthesis.com/products/xsd/ hth, -boris -- Boris Kolpackov Code Synthesis Tools CC http://www.codesynthesis.com Open-Source, Cross-Platform C++ XML Data Binding --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
