Thanks Frank. See responses inline.
-----Original Message----- From: Frank Stearns [mailto:fra...@fsatools.com] Sent: Wednesday, October 02, 2013 8:29 PM To: VLM TechSubs Cc: 'Frank Stearns'; 'Framers' Subject: RE: FM 8: Index marker syntax On Wed, 2 Oct 2013, VLM TechSubs wrote: > 1. I now see that even a simple marker entry "<Emphasis>Eisphora" > has this problem. It appears in the "correct" location alphabetically, > it just looks like > > . . . . . . 1 > > 2. The original files were created in Word, but with no indexing. I > put in 100% of these markers by hand, by myself, this week, in > FrameMaker 8. In other words, they "shouldn't" be affected by Word or > the like. FYI, the ToC works just fine. > > 3. I just did the following: > > 3.1 Shut Frame. > 3.2 Relaunched Frame. > 3.3 Opened book file. > 3.4 Opened all files in book. Got some font and xref error messages, clicked > through those. > 3.5 Highlighted, then saved, all files in book. > 3.6 Generated ToC and index. > 3.7 Saved all files in book. > 3.8 Opened index. > 3.9 Found one of the offending entries, jumped to its source. It was a > compound marker (2 entries in one marker). > 3.10 I copied one of the entries for reference, clicked in the > document a few words away, and manually typed in the text for the > marker "America:Income tax of 1894", just like that. > 3.11 Went to book file, all files selected, save all files in book, > then generate. > 3.12 Entry appears in the correct location, without any text only page > #, and when I jump from that page number it takes me to the new marker > I just created. > 3.13 So I jump there, notice that the original marker is still in > place, delete it, returned to 3.7 or so, repeated the whole process, > same result. > 3.14 Saved the index as mif, closed it, deleted the .fm version, > reopened from mif, saved as FM 8, went to book, save all files in > book, generate, save all files in book, jump to index, and voila! > Same result as before. > > Sigh ... > > BTW, between steps, I recall breathing. :) > > Frank, I am most willing and eager to try your step #4, but I've > forgotten how to open the mif as text in Frame, after saving it. I know there's a way to do that, but can't recall it. We've always used a programmer's plain text editor -- even notepad or wordpad will work fine. Avoids trying to get Frame to play in a way it doesn't "want" to. :) This is all pretty nutty, that's for sure. A few more random thoughts... 1. The Word source still makes me cautious. If you highlight text to create a marker, or even were to use our IXgen tool to create markers, there can still be those nasty non-printing characters that would be picked up from body text and transferred into a new marker and cause problems. Frank, you are correct, at least inasmuch as I've removed hundreds of nonprinting characters from these documents. But almost all were at the end of paragraphs, so whenever I've placed a marker at EOP, I've carefully removed any nonprinting characters that may have been present. 1A. When you hand typed a marker, was the entry field completely clear? (Those non-printing chars can be insideous. Click next to one in the body that you can't see and the marker dialog autofill might pick it up, and you'd never know it. Most of the time, I typed it all in character by character. At times, I did begin with a highlighted word or two. But many markers created in this way work fine, and some of the failing markers were definitely created by hand typing, from scratch. I suppose I could take this entire document set out to a text editor, do something there (not sure quite what), then bring it all back into FM and start from scratch. Ugh . but it's a possibility. 2. Forgot to verify the obvious: most of the markers are working correctly, and these are just isolated cases? Or, are you seeing the same thing happen any time you have an entry with a second level and/or font change? Yes, majority of markers appear to be working fine. It happens on single, double, and triple level entries. It happens with or without a font change. 3. Have you done a cold restart of your system? Shut it down completely, power supply switch OFF for a few minutes, then a reboot? I know, it's crazy, however.... Cold restart, yes, long enough to go to the grocery. :) Could do it again I suppose. I agree, crazy and however. :) 4. In the IX file you generate, have you examined the reference pages that set up your entry formats? Are all the components there? Have you examined the entry para formats for anything odd? If you move the cursor through the entries in the main flow, do you notice if in the problem entries the para or char tag name suddenly becomes starred, indicating a deviation from the catalog settings? _____ >From the reference page: 1, 2-3 Level4IOM Level3IOM Level2IOM Symbols[\ ];Numerics[0];A;B;C;D;E;F;G;H;I;J;K;L;M;N;O;P;Q;R;S;T;U;V;W;X;Y;Z <$symbols><$numerics><$alphabetics> -_-- Level1IOM openObjectId <$relfilename>:<$ObjectType> <$ObjectId> <$pagenum> _____ Moving the cursor through the entire main flow shows no paratag or chartag anomalies. On the reference page, however, such anomalies do appear. Could that mean something? I did find level 4 strange, as I did not intentionally create any 4-level entries. When I looked, I found that I typed a comma instead of a semicolon in exactly one marker. So that problem was easy to fix, I just haven't removed the extra line from the reference page. Finally, I have a very fuzzy memory that FM8 had introduced some bugs in marker handling... Anyone else remember this? Do you have the lastest version of FM8? I have version 8.0p266, and no memory WRT FM8 indexing bugs. Frank, do you have or know of any special diagnostic tool(s) I might use here? Continued good luck with it!! Frank Stearns Associates | makers of IXgen(tm) for FrameMaker(r) *** IXGEN 7-10 for FRAMEMAKER 10/11 IS HERE! *** <mailto:franks at fsatools.com> franks at fsatools.com < <mailto:franks at fsatools.com> mailto:franks at fsatools.com> TOLL FREE Voice (USA and Canada): 800-567-6421 USA Voice: 435-275-8981 <http://www.fsatools.com> http://www.fsatools.com -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.frameusers.com/pipermail/framers/attachments/20131002/8f518f61/attachment.html>