I went ahead with the plan I outlined earlier and it seems to run- I can make text visible or invisible as desired by using dvi specials. I added one cpp file that registers a special from a global variable during ctor execution and a few header files including a "c" file included in an exsting "c" file to avoid adding to project :) To avoid dealing with interface between ev-view and dviread, I send mouse coords out to datagram and requests back to reload the page with new parameters. It more or less works, blocks come and go, except that the thing I point at is not what gets modified. I'm trying to compare event->x + view->scroll_x to dvi->pos.hh and there is maybe an offset or offset and scale issue. Where is an example of the right conversion? The datagram thing sounds dumb but it does create flexibility and avoids having the deal with interfaces right now. I can even pass pointers back and forth. Thanks. -- mike marchywka 306 charles cox canton GA 30115 USA, Earth [email protected] 404-788-1216 ORCID: 0000-0001-9237-455X _______________________________________________ evince-list mailing list [email protected] https://mail.gnome.org/mailman/listinfo/evince-list
[evince] translating event->x and event->y into dvi->pos.hh and dvi->pos.vv coords
Mike Marchywka via evince-list Sat, 12 Oct 2019 13:22:18 -0700
