not to question your method, I'd just like to know better the requirements...... line numbers change (a lot) between commits. They'll be valid only in the context of the commit, i.e. they will be exact pointers only referring to the "past" status of the source file.
What will happen is head^2 1. lorem ipsum 1.dolor sit amet 2. dolor sit amet 2. head^1 1. dolor sit amet 1.dolor sit amet 2. 2.consectetur adipisicing elit head 1.dolor sit amet 1.haha, big joke 2.consectur adipiscing elit 2.removed alltogether If I were a translator at head^2, I'd just watch for new "sections" and removed ones, searching by context (that's why diffbook outputs some lines before and after each changed line) .... then I'll insert new "sections" translating what's necessary. At that point linenumbers are just unhelpful (unless you managed to keep linenumbers in sync with the original source, hard to believe ..... but in that case either, you want to translate from head^2 to head, possibly skipping the "partials" between head^2 --> head^1 and head^1 --> head ) tl;dr : How can line numbers be helpful on a commit-by-commit base? On Friday, March 29, 2013 11:02:12 PM UTC+1, Alan Etkin wrote: > > you mean both on the original and on the new "stacks" ? >> > > I suppose is enough if they are appended to the right side stack. That's > what I'd use for browsing the places to modify in the chapter files. > > -- --- You received this message because you are subscribed to the Google Groups "web2py-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
