Hey mighty scripters. I think I sent this one late last week, but didn't get any nibbles on it so am trying again. if any kind soul could give me a hand instituting this one, it'd be heaps appreciated. Also, may I add one more queery to the top? What's the bestest function for announcing only the text of a certain color, from a string? i.e. I do a couple of getLines() and concatinate them into a string, in order to retain tab Characters which I need to remain so I can replace them with commas for pauses, and then I wanna say, let that string only equal text that's xx on YY color. and then just speak that data out of the string? I know you can use special optional color parameters on GetTextInRects, but in this case I can't use that, because it strips out the Tab Characters automatically, and I need functions that don't do that? Thanks for any tips ASAP on that one, and if I may, here comes the repeat of last weeks post, which I'm still very keen to implement on the last day of the job! ah, feature creep. gotta love it! Sent: Friday, April 17, 2009 10:49 AM Subject: [jawsscripts] instituting a find command across multiple pages of data? oh mighty > Scripters. I have only two days left here at the office of the project at > which I've been working for the past xx weeks, and I'd reeeeally like to > implement one more feature that I was wondering if someone could help me > with a bit? > I Got a situation where searching for a street number across multiple > screens/pages would be rather desirable to have a computer-based method of > doing, in that the sightlings can far more easily glance down a list of > records and > just manually flip pages till they find it, than the vip can. > > So, what I was wanting to do, was implement some iteration of jawsFind, or > FindString command, that allowed me to: > A. have the user hit control+f and bring up a box asking what to search for, > as in the jawsFind box or something equivalent. > B. have them type that string in to an edit box field for reception of it, and hit enter. > C. for the script to look on the current screen, if not found, flip to the > next page, pause a bit to insure settling, then look on that screen. if > not > there, > flip to the next page and so on, until found. > > obviously implementing some limit/safety while count thinggy to stop at a > certain number of pages etc. > > Then, if/when found, take the jaws cursor to the beginning of that string > onscreen, then flip back to the left a certain distance and click in the > spot > which will immediately bring focus to that record for the user, such that > my > onFocus frame commands then kick in and read the appropriate data from it. > > Could such be done? > > I've started having a bit of a mess with this myself, but, ... well > although I can do all the end part ok myself, I've got such lame knowledge, that I'm even unsure how to stop the normativ jawsfind > box, from wanting to pop up another message/ok box thinggy, as an error on > the first page when it can't find the string on that screen. As of course > this is what jawsFind function is normatively supposed to do? > > But then wehn I thought about seeking to utilize the standard FindString > function, I realized I was additionally so knowledge-lame that I had no idea how to make > code that could prompt the user for the string to insert into that second > parameter of the FindString function, to even start the whole ball > rolling! I have never made jaws pop up boxes for reception of user input before? > If any kind soul could assist with anything here, be very much > appreciated! > > I was initially gunna do this in virtual buffer land, loading maybe 20 > pages > of data or something into the virtual viewer, and seeking to keep track of > what page of data was what, utilizing some kind of variable or something. > then if found in virtual buffer land search, which I knew would be faster > than > the more "live" method I've proposed above, and then try and go back to > the > live option and find the page, and then do the jawsFind onscreen. But I'm > thinking now that the live method I've proposed above would be just a lot > simpler and more straight ahead, not worrying about virtual buffer issues > etc. > > As I say, if anyone could give me any tips on how I might implement, this > would be just great! > > thanks. > > geoff c. > > __________ > Visit and contribute to The JAWS Script Repository http://jawsscripts.com > > View the list's information and change your settings at > http://www.freelists.org/list/jawsscripts > > __________ Visit and contribute to The JAWS Script Repository http://jawsscripts.com View the list's information and change your settings at http://www.freelists.org/list/jawsscripts