Inject and index a single url with Nutch

Question

Hello;

I want to inject a single url to the crawldb as a string not a urlDir, I'm thinking in add a modified method of the Injector.inject that take the url as a string parameter, but I cant inject the string url in the crawldb; I guess the current injector using the fileInput.. from hadoop.

how can I do this ?

and I test to crawl url in a txt file using the urlDir, with depth equals to one but when I change the url to crawl another one, it gives me an error about "segmants", and it try to fetch the old url larger depth, I guess this is about marging things, and I am trying to handle it. But if any resource helps, I will be thankful :)

Thanks!