Jump to content


Photo

Index Problems in RTF Files Revisited


  • Please log in to reply
24 replies to this topic

#1 Jerry Bryan

Jerry Bryan

    Advanced Member

  • Members
  • PipPipPip
  • 3590 posts

Posted 26 September 2010 - 11:23 AM

It is two weeks from the main family reunion for which I produce a narrative report each year, and I am having my annual battle with RM with respect to getting the report to look the way I want it to look. With FO, RM1, RM2, and RM3, I always won the battle and produced a good looking report. Last year with RM4, I would describe the battle as sort of a draw. The report didn't look perfect, but it was acceptable. This year with RM4, I think RM4 is winning and I'm losing.

I can't get the indexes to work correctly with an RTF file this year (about which more later), so my plan is to print directly from RM4. The report will look sort of ok, but it will still have three problems. 1) I won't have an opportunity to clean up a variety of white space problems in the standard RM4 formatting. The most egregious problem is a blank at the beginning the first line for a fact after a hard return in the fact note for the preceding fact. 2) RM4 does not support guttering for even/odd pages when using duplex printing (printing both front and back). With an RTF file, I could achieve proper guttering by printing with Microsoft Word. 3) I won't be able to keep a copy of the print file for my records, since there won't be a print file. Well, in addition to trying RTF, I also tried to make a PDF file. But the PDF file appears as gibberish in Acrobat Reader. There is another thread about the PDF problem in progress as we speak.

So, back to the RTF index problem. The problem may well be more of a Microsoft Word problem than an RM4 problem. However, I can make my own Microsoft Word document using what appears to be the same index entries that RM4 is creating. And my manually created Microsoft Word file has multiple indexes that render perfectly. So I don't understand why the RTF file rendered by RM4 doesn't render correctly. I could decipher the RTF file created by FO, RM1, RM2, and RM3. But the RTF file created by RM4 is using style sheets in a manner that I don't understand, and I wonder if the problem isn't in the style sheets somehow or other. Here are my symptoms.

a. Upon creating the RTF file and opening it with Word, the Name Index and the Place Index both have just one column. In RM4, I specified three columns for the Name Index and two columns for the Place Index, and both indexes have the correct number of columns when looked at from within RM4. Also, upon opening the RTF file in Word, every single page number in both indexes is page 3.

b. I followed the procedure to regenerate the indexes as per the RM4 Knowledge Base article at http://support.roots...BSearchID=42517 After doing so, the Name Index looks ok, and the Place Index doesn't look ok. Well, in a strange sense, the Place Index contains correct information, but it doesn't "look ok". Here's an sample from the Place Index that exemplifies the problem.

Tennessee

[Knox County] 18, 30, 31, 34, 35, 37, 38
[Union County] 9, 40
Anderson County

Claxton

Tennessee Valley Memory Gardens 24, 25

Clinton 28, 39, 44, 60
Edgemoor 3, 7, 13, 19, 23
Oliver Springs 5, 10, 22, 26
Scarbrough 12, 13, 25, 28, 32, 42, 43

Anderson County 1, 2, 3
Anderson County 4
Anderson County 4
Anderson County 5
Anderson County 5
Anderson County 5
Anderson County 5
Anderson County 5
Anderson County 5
Anderson County 5
Anderson County 5
Anderson County 5
Anderson County 6
Anderson County 6
(etc.)


There are two problems, one which is very visible and one which is not. The very visible problem is all those duplicate entries for Anderson County that are not merged together properly. These are the cases where Anderson County is not further subdivided. The Anderson County entries look fine in the version of the report generated directly by RM4. They are obviously not correct in the version generated by Microsoft Word. That's why I'm going to print directly from RM4 this year. And I repeat that I cannot duplicate the problem when I create a report totally by hand in Word and index it totally by hand in Word. If I do so, the index looks fine (and yes when I test by hand, I do make multiple indexes).

The other problem is not so visible. There are a number of cemeteries that should appear in the index. They should appear under Tennessee, Anderson County, cemetery_name. One cemetery does appear correctly, namely Tennessee Valley Memory Gardens. It appears in the format Tennessee, Anderson County, community_name, cemetery_name, where the community_name is Claxton. But the rest of the cemeteries don't appear at all in the index, not even in the version of the report generated by RM4. So this problem can't possibly be blamed on the RTF format or on Microsoft Word. The way I'm entering place names and place details, either "Anderson County, Tennessee" or "community_name, Anderson County, Tennessee" appears in the place name, and cemetery_name appears in the place details. Then in the index, the names are reversed to put Tennessee first, Anderson County second, etc. In a more urban area, the same problem exists of cemetery names not appearing in the index except that community_name would typically be city_name instead. When reversing the names for the purpose of the index, RM4 is apparently getting confused when the number of levels is not always the same.

Jerry Bryan

#2 Jerry Bryan

Jerry Bryan

    Advanced Member

  • Members
  • PipPipPip
  • 3590 posts

Posted 26 September 2010 - 12:07 PM

I have one correction to report. I thought I had split all my cemetery names into the Place Details field. However, the two references to the Tennessee Valley Memory Gardens were cases where I had failed to find the cemetery name in the Place Name and split it into the Place Details field. Upon making that correction in my database, that strangeness in my Place Index went away. Indeed, it appears that the Place Details do not appear in the Place Index at all. I'm going to have to think about whether that is good or bad. I kind of liked having cemetery names in the Place Index.

Upon making this correction to my database, all the other problems I mentioned for RTF files remain. But the apparent inconsistency in reports generated directly from RM4 no longer exits.

Jerry Bryan

#3 Nettie

Nettie

    Advanced Member

  • Members
  • PipPipPip
  • 1647 posts

Posted 26 September 2010 - 12:12 PM

Jerry

I was able to reproduce the pdf file with a narrative report of 6 gen that matched what was being given in RM4 screen before saving to a pdf file.

I transferred the same file to MSWord2003 and yes the two indexes were in 1 column and was able to make then 2 columns. I agree the Place Index does not catch all the cemeteries that were entered into the database. Some of them but not all. Don't know why. One issue I always have when the transfer from rtf to a Word document is that in the RM4 programing they are using two different types of symbols for end of paragraph on most of them they look like a backwards 7. But the other one looks like a tab marking with a left hand curve. Why 2? :huh: I usually do in Word a Cont+H and down on lower choices is Specials click on that and select Manual line break and change all of them to ^p, then the changes actually work so much easier. I do not know why they are using 'Manual Line Break' unless certain software requires it.

Genealogy:
"I work on genealogy only on days that end in "Y"." [Grin!!!]
from www.GenealogyDaily.com.
"Documentation....The hardest part of genealogy"
"Genealogy is like Hide & Seek: They Hide & I Seek!"
" Genealogists: People helping people.....that's what it's all about!"
from http://www.rootsweb....nry/gentags.htm
Using FO and RM since FO2.0 


#4 Nettie

Nettie

    Advanced Member

  • Members
  • PipPipPip
  • 1647 posts

Posted 26 September 2010 - 12:21 PM

You're right :) the place details in the place list does not work, I thought that was one of the fixes a few revisions back.

Genealogy:
"I work on genealogy only on days that end in "Y"." [Grin!!!]
from www.GenealogyDaily.com.
"Documentation....The hardest part of genealogy"
"Genealogy is like Hide & Seek: They Hide & I Seek!"
" Genealogists: People helping people.....that's what it's all about!"
from http://www.rootsweb....nry/gentags.htm
Using FO and RM since FO2.0 


#5 Renee Zamora

Renee Zamora

    Advanced Member

  • Support
  • PipPipPip
  • 8470 posts

Posted 27 September 2010 - 10:43 AM

You're right :) the place details in the place list does not work, I thought that was one of the fixes a few revisions back.

There is an enhancement request in our tracking system to print place details in the place list index on reports. The fix you saw was probably this one - New: Place list report can now include place details VERSION 4.0.8.4 (5/6/2010). The place details are only viewable on the Place list report itself, not in the indexes on various reports.
Renee
RootsMagic

#6 Jerry Bryan

Jerry Bryan

    Advanced Member

  • Members
  • PipPipPip
  • 3590 posts

Posted 27 September 2010 - 02:58 PM

Tennessee

]
Anderson County 1, 2, 3
Anderson County 4
Anderson County 4
Anderson County 5
Anderson County 5
Anderson County 5
Anderson County 5
Anderson County 5
Anderson County 5
Anderson County 5
Anderson County 5
Anderson County 5
Anderson County 6
Anderson County 6
(etc.)


I think I have solved this piece of the problem where the duplicate entries for Anderson County are not properly merged in the index. I believe there is an error in the XE entries produced by RM4 for the index. The error is systematic enough that I can fix it by doing a global replace in Word. A ticket needs to be opened and a bug needs to be squashed, but I can work around the bug for now. In order to fix it, while I'm in Microsoft Word I have to make the XE entries visible. To do so, I just click the little paragraph symbol (the funny looking P) in the ribbon at the top of the Microsoft Word 2007 screen. I then have to use global replace to remove a blank that shouldn't be there in all the XE entries. Then I have to make the XE entries invisible again before generating the index, else many of the pages numbers will be incorrect in the index.

This is a tricky problem. To describe it properly, I need to use a character other than a blank to represent a blank. I will use a b with strike through, e.g. b. Places names appear in the XE entries as quoted strings. If you reverse the names, for Anderson County, Tennessee it is the case that RM4 produces the quoted string "Tennessee:AndersonbCountyb" in the XE entry when it should in fact be producing "Tennessee:AndersonbCounty" in the XE entry. The fix is as simple as that.

The b that shouldn't be there gets Microsoft Word very confused. Well, under most circumstances, Microsoft wouldn't get confused. It gets confused because I have places such as Claxton, Anderson County, Tennessee and Scarbrough, Anderson County, Tennessee along with just plain old Anderson County, Tennessee. Here's what the quoted strings in the XE entries for these three locations look like.

"Tennessee:AndersonbCounty:Claxtonb"
"Tennessee:AndersonbCounty:Scarbroughb"
"Tennessee:AndersonbCountyb"

The b really shouldn't be at the end of Claxton and Scarbrough in the first two entries, respectively, but it doesn't really hurt anything at that point. The killer is the b at the end of County in the third entry. Because now, Tennessee:AndersonbCountyb doesn't match the Tennessee:AndersonbCounty part of Tennessee:AndersonbCounty:Claxtonb and Tennessee:AndersonbCounty:Scarbroughb, and Word gets very confused. Once it gets confused, it stays confused. The mismatch in the strings is very minor - just the trailing space. But that's all it takes to cause the strings to fail to match.

So for now, I can fix the error in my RTF file and print successfully with Word so I will have proper gutter control. For the future, the bug needs to be squashed.

Jerry Bryan

#7 Nettie

Nettie

    Advanced Member

  • Members
  • PipPipPip
  • 1647 posts

Posted 27 September 2010 - 03:23 PM

Thanks for sharing your find. Will keep my eyes open for the issue in my items/stuff :)

Thanks goodness for the forum.

Genealogy:
"I work on genealogy only on days that end in "Y"." [Grin!!!]
from www.GenealogyDaily.com.
"Documentation....The hardest part of genealogy"
"Genealogy is like Hide & Seek: They Hide & I Seek!"
" Genealogists: People helping people.....that's what it's all about!"
from http://www.rootsweb....nry/gentags.htm
Using FO and RM since FO2.0 


#8 TomH

TomH

    Advanced Member

  • Members
  • PipPipPip
  • 6254 posts

Posted 27 September 2010 - 08:16 PM

Great sleuthing, Jerry! Now if I could find a way to get the RM4 RTF working in Open Office Write...

Tom user of RM7550 FTM2017 Ancestry.ca FamilySearch.org FindMyPast.com
SQLite_Tools_For_Roots_Magic_in_PR_Celti wiki, exploiting the database in special ways >>> RMtrix-tiny.png app, a bundle of RootsMagic utilities.


#9 Jerry Bryan

Jerry Bryan

    Advanced Member

  • Members
  • PipPipPip
  • 3590 posts

Posted 28 September 2010 - 10:33 AM

This is a tricky problem. To describe it properly, I need to use a character other than a blank to represent a blank. I will use a b with strike through, e.g. b. Places names appear in the XE entries as quoted strings. If you reverse the names, for Anderson County, Tennessee it is the case that RM4 produces the quoted string "Tennessee:AndersonbCountyb" in the XE entry when it should in fact be producing "Tennessee:AndersonbCounty" in the XE entry. The fix is as simple as that.


I should probably add the following clarification. The problem I'm describing occurs with lots of place names in my report, not just Anderson County, Tennessee. The temporary bypass that will take care of all the place names in the report is to use Word to do the following.

1. Show the text in the XE entries by clicking the paragraph symbol on the Word 2007 ribbon.
2. Do a global replace of the string ' " \f B' with the string '" \f B' (without the single quotes). This process will get rid of all the trailing blanks in the place index that are causing the problem. The trailing blanks are immediately in front of a double quote mark, which is why this replace process will solve the problem, pending a fix in RM4 itself. It probably wouldn't hurt also to do the same replacement in the person index (use \f A rather than \f B for the person index), but the trailing blanks in the XE entries for the person index are apparently not causing any formatting problems in the index.
3. Hide the text in the XE entries by clicking the paragraph symbol on the Word 2007 ribbon.
4. Generate the indexes as you normally would.

Jerry Bryan

#10 Renee Zamora

Renee Zamora

    Advanced Member

  • Support
  • PipPipPip
  • 8470 posts

Posted 28 September 2010 - 01:05 PM

I should probably add the following clarification. The problem I'm describing occurs with lots of place names in my report, not just Anderson County, Tennessee. The temporary bypass that will take care of all the place names in the report is to use Word to do the following.

1. Show the text in the XE entries by clicking the paragraph symbol on the Word 2007 ribbon.
2. Do a global replace of the string ' " \f B' with the string '" \f B' (without the single quotes). This process will get rid of all the trailing blanks in the place index that are causing the problem. The trailing blanks are immediately in front of a double quote mark, which is why this replace process will solve the problem, pending a fix in RM4 itself. It probably wouldn't hurt also to do the same replacement in the person index (use \f A rather than \f B for the person index), but the trailing blanks in the XE entries for the person index are apparently not causing any formatting problems in the index.
3. Hide the text in the XE entries by clicking the paragraph symbol on the Word 2007 ribbon.
4. Generate the indexes as you normally would.

Jerry Bryan

I would like to add this issue in our tracking system, but I can't get my index to create the duplicate counties like you have. Would you be willing to submit a ticket on this, with the resolution and include a backup copy of your database for testing?

Submit a Ticket
http://support.roots...in/Default.aspx
Renee
RootsMagic

#11 Jerry Bryan

Jerry Bryan

    Advanced Member

  • Members
  • PipPipPip
  • 3590 posts

Posted 28 September 2010 - 06:24 PM

I would like to add this issue in our tracking system, but I can't get my index to create the duplicate counties like you have. Would you be willing to submit a ticket on this, with the resolution and include a backup copy of your database for testing?

Submit a Ticket
http://support.roots...in/Default.aspx


The Ticket has been successfully submitted.
Ticket Number 2D2-14353CF4-5B5F

Jerry

P.S. I created a little 2 generation database containing 18 people, which was sufficient to exemplify the problem. I submitted both a copy of a backup file for the little database, and also an RTF file that has the bad Place Index. I created the database with drag and drop. All my facts have notes, even if the note is blank, and all my fact notes end with two carriage returns to produce a blank line between each fact in the report. When I did the drag and drop, the fact notes in the new database all ended with just one carriage return. So the last carriage return was omitted from every note as a part of the drag and drop. That's a pretty serious bug that should be trivial to reproduce and that badly needs to be squashed.

#12 Renee Zamora

Renee Zamora

    Advanced Member

  • Support
  • PipPipPip
  • 8470 posts

Posted 29 September 2010 - 10:03 AM

The Ticket has been successfully submitted.
Ticket Number 2D2-14353CF4-5B5F

Jerry

P.S. I created a little 2 generation database containing 18 people, which was sufficient to exemplify the problem. I submitted both a copy of a backup file for the little database, and also an RTF file that has the bad Place Index. I created the database with drag and drop. All my facts have notes, even if the note is blank, and all my fact notes end with two carriage returns to produce a blank line between each fact in the report. When I did the drag and drop, the fact notes in the new database all ended with just one carriage return. So the last carriage return was omitted from every note as a part of the drag and drop. That's a pretty serious bug that should be trivial to reproduce and that badly needs to be squashed.

I see your ticket and will start working on it. Thanks for the additional information on the carriage returns. I am going to check if that has been reported before or not.
Renee
RootsMagic

#13 Renee Zamora

Renee Zamora

    Advanced Member

  • Support
  • PipPipPip
  • 8470 posts

Posted 29 September 2010 - 12:44 PM

Jerry,
Looking at the RTF you sent shows the duplicate county names. Looking at one created from your backup everything is correct. I know you used the Register Report. Do you recall any other settings you may have used?
Renee
RootsMagic

#14 Renee Zamora

Renee Zamora

    Advanced Member

  • Support
  • PipPipPip
  • 8470 posts

Posted 29 September 2010 - 01:39 PM

1) I won't have an opportunity to clean up a variety of white space problems in the standard RM4 formatting. The most egregious problem is a blank at the beginning the first line for a fact after a hard return in the fact note for the preceding fact.

The spacing issues have previously been submitted to our tracking system for resolution.

2) RM4 does not support guttering for even/odd pages when using duplex printing (printing both front and back). With an RTF file, I could achieve proper guttering by printing with Microsoft Word.

If you are printing directly from RM4 you can set the printer options to do duplex printing. (Click the Print icon and then go into Properties.) I have added an enhancement request to allow saving of printer settings to the RTF or PDF created.

Indeed, it appears that the Place Details do not appear in the Place Index at all. I'm going to have to think about whether that is good or bad. I kind of liked having cemetery names in the Place Index.

This enhancement request is also in our tracking system.

So the last carriage return was omitted from every note as a part of the drag and drop.

This has also been previously report and is in our tracking system.

The issue with the PDF has been commented on in the other thread. So I think this covers everything that needs to be addressed.
Renee
RootsMagic

#15 Jerry Bryan

Jerry Bryan

    Advanced Member

  • Members
  • PipPipPip
  • 3590 posts

Posted 29 September 2010 - 04:29 PM

Jerry,
Looking at the RTF you sent shows the duplicate county names. Looking at one created from your backup everything is correct. I know you used the Register Report. Do you recall any other settings you may have used?


The Register report is the only one I did myself.

Did you actually create an RTF file from my backup and generate the indexes in Word? If you look at the Register report from within RM4 itself, the index is ok. The failure is only after you save the report as an RTF and open the RTF in Word.

Jerry

#16 Jerry Bryan

Jerry Bryan

    Advanced Member

  • Members
  • PipPipPip
  • 3590 posts

Posted 29 September 2010 - 04:54 PM

If you are printing directly from RM4 you can set the printer options to do duplex printing. (Click the Print icon and then go into Properties.) I have added an enhancement request to allow saving of printer settings to the RTF or PDF created.


Renee, if I click Print and then Properties, I have no duplex option. That's for the simple reason that my printer will not do duplex printing. I print a master report one-sided and take it to a print shop to get the double sided copies made.

But duplex printing really isn't the issue. The issue is gutter control. In order to bind the reports properly after they are printed as double sided copies, the inside margin has to be larger than the outside margin. For the odd pages, the inside margin is the left margin and the outside margin is the right margin. For the even pages, the inside margin is the right margin and the outside margin is the left margin. That's what gutter control does.

I print with a 0.75 inch margin plus a 0.5 inch gutter. That makes the inside margin really be 1.25 inches and the outside margin really be 0.75 inches. Gutter control adjusts the right and left margins to be appropriate for even and odd pages. That's one of several reasons I print from Word rather than from RM4. Word supports gutter control and RM4 does not. But even if RM4 had gutter control, I would still want to print from Word rather than from RM4 for other reasons.

Jerry

#17 Jerry Bryan

Jerry Bryan

    Advanced Member

  • Members
  • PipPipPip
  • 3590 posts

Posted 29 September 2010 - 05:37 PM

I know you used the Register Report. Do you recall any other settings you may have used?


I tried all five kinds of narrative reports, a Pedigree report, and a family group sheet - specifying both a Name Index and a Place Index for each report. Every report I ran had the same problem in the Place Index of failing to merge duplicate place entries. I didn't try every possible report under Lists, but I'm confident they will have the same problem as well.

Jerry

#18 Nettie

Nettie

    Advanced Member

  • Members
  • PipPipPip
  • 1647 posts

Posted 29 September 2010 - 06:30 PM

Jerry
I know this could be tedious, but....

I have two printers [Canon MP850 & Canon IP80] and the Canon IP80 that goes with me is for my laptop does not duplex.
But if you are setting up word for the gutters being the correct side of a odd number page and correct side of an even number page. You can do one page at a time, after it comes out put it back in so it prints on the duplex/back side. What I did all winter while at my daughters. Duplexing saves paper = green. It is tedious, but had a grandson that loved to push the button for printing. He was 2 and got it right every time.

Other choice set your word document up with the right gutters/settings and put it on a flash drive or a CD, take your copy to a place like Kinkos [MN co.] print it and then use it for making your multi copies.

Just another suggestion. Yes it is great to find out why things are not working, but sounds like you are on a schedule to get this done. Good Luck :)

Genealogy:
"I work on genealogy only on days that end in "Y"." [Grin!!!]
from www.GenealogyDaily.com.
"Documentation....The hardest part of genealogy"
"Genealogy is like Hide & Seek: They Hide & I Seek!"
" Genealogists: People helping people.....that's what it's all about!"
from http://www.rootsweb....nry/gentags.htm
Using FO and RM since FO2.0 


#19 Jerry Bryan

Jerry Bryan

    Advanced Member

  • Members
  • PipPipPip
  • 3590 posts

Posted 29 September 2010 - 08:04 PM

But if you are setting up word for the gutters being the correct side of a odd number page and correct side of an even number page. You can do one page at a time, after it comes out put it back in so it prints on the duplex/back side.


I've done what you suggest bunches of times in the past. But I'm up to needing 30 copies of a 90 page report - just too tedious to print one page at a time. That's why I now go to a print shop to get my reports printed front and back. Well, what I've really done in the past is to print all the even pages, turn them over and then print all the odd pages on the back. It's not quite so tedious as one page at a time, but still pretty tedious. The problem is that the printer jams pretty badly when paper is fed through a second time, even if it never jams on the first pass. Going through the printer bends the paper a little, changes the moisture content of the paper a little, and builds up static electricity. Even waiting several hours between the first pass and the second pass to let the paper readjust doesn't help very much.

Jerry

#20 Jerry Bryan

Jerry Bryan

    Advanced Member

  • Members
  • PipPipPip
  • 3590 posts

Posted 30 September 2010 - 07:12 AM

Jerry,
Looking at the RTF you sent shows the duplicate county names. Looking at one created from your backup everything is correct. I know you used the Register Report. Do you recall any other settings you may have used?


Renee, here's one more thing about being able to reproduce the problem. Even after creating the RTF file and opening it with Microsoft Word, the problem may not really be visible immediately. To make it visible, you probably have to do a CTL-A to select all the text in the document and then do an F9 to rebuild the indexes.

Jerry