On 2014-04-29, Bing Lu wrote:
Also I were to parse the DFDL using IBM DFDL code, is there a setting that I can call to do partial parsing? And when parsing happens, does it stop on the first parsing error or it tries to parse as much as possible to include all the errors encountered?
I too would like to see both the proper upper and lower parsing complexity of the language proven. Not asymptotically, but over the minor oh. After that, I'd really like to see the best known constants as of last week, and the whole rationale why the language took this precise, hypermicromanaged route where you have to ask about the existence of a formal grammar in the first place. Seriously, I've pretty much been unemployed for two years after you already did two years of hard work. I'm still not seeing the promise even in IBM's code. What is this, another SGML?!? When is this thing going to parse my favourite, random format, out of the box? -- Sampo Syreeni, aka decoy - decoy@iki.fi, http://decoy.iki.fi/front +358-40-3255353, 025E D175 ABE5 027C 9494 EEB0 E090 8BA9 0509 85C2