Jump to content
PiercysEtc

Indexing Journal Report

Recommended Posts

I'm running TMG 7.04 under Win XP SP3 and exporting a Journal report into Word 2000 with two of TMG's standard automatic indices, People and Places, in the report.

 

The People index works fine, outputting conventionally:

..Surname

....Forename

 

However, problems with the Places index.

 

I have not found a way to order the Place index into the conventional format:

..Country

....State, City, etc...

or even:

..Country

....State

......City, etc...

without disordering the sentence place structure.

 

For example, using Short Place Field in both Report Options > Indexes > Places and Report Options > Places > Places, and setting the Short Place Template to:

<[L6], ><[L5], ><[L4], ><[L3], ><[L2]>

I can get the index to order in the conventional format above. However, the variable [L] in the sentence now resolves to the same order, Country, State, etc; the opposite of what convention expects.

 

I've worked through Lee Hoffman's 'Getting the Most Out of TMG' but he only has a small section on Place Indexing and doesn't seem to cover this issue specifically. I've searched Rootsweb's TMG list and posted there but no luck.

 

I must have missed something - can anyone out there who has successfully implemented an automatic Place Index please tell me what I've missed?

 

Thanks in advance.

 

Tom Piercy

Edited by piercyetc

Share this post


Link to post
Share on other sites

The point missed is that sentence variables are designed to control sentence output. If you use a short place template as you describe, the place part output will follow that order in sentence output as you discovered.

Share this post


Link to post
Share on other sites

In the TMG Help in the topic "Report Options: Indexes" under the section "Creating Index Entries" it mentions the capability to add your own index entries to sentences if the ones automatically generated are not sufficient for your purposes.

 

See: [iNDEX:]index:val1,val2,val3[:INDEX]

 

I have not used them myself, but you might try various forms in a few sentences and test if they give you what you want.

Share this post


Link to post
Share on other sites
The point missed is that sentence variables are designed to control sentence output. If you use a short place template as you describe, the place part output will follow that order in sentence output as you discovered.

Thanks, Jim, for your very prompt reply. I was trying to say that this (inappropriate 'fudge', as you rightly note) is the only way I have found to order the index into the conventional format. However, the short place template then does what it was designed to do, and orders the place output in the sentence - not what I want!

 

So, to clarify my query, what is the correct way to order the Places index into the conventional format as below?

..Country

....State, City, etc...

 

Thanks again

 

Tom

Edited by PiercysEtc

Share this post


Link to post
Share on other sites
In the TMG Help in the topic "Report Options: Indexes" under the section "Creating Index Entries" it mentions the capability to add your own index entries to sentences if the ones automatically generated are not sufficient for your purposes.

 

See: [iNDEX:]index:val1,val2,val3[:INDEX]

 

I have not used them myself, but you might try various forms in a few sentences and test if they give you what you want.

 

Thanks Michael. I had looked at this and if all else fails then I guess this might be a way to go. I don't fancy the idea of indexing over 6,000 places by hand when TMG can do it automatically, though. The Names index works excellently so I'm sure that the Places index will give equally good results if I can just suss out the right configuration :wub: .

 

Best wishes

 

Tom

Share this post


Link to post
Share on other sites

I have not tried this, but you might test using the location variables rather than text in a modification to the Tag Type sentences. Then you don't have to deal with individual places. This is a guess, but perhaps something like:

[iNDEX:]places:[L6],[L5],[L4][:INDEX]

added to sentence templates that you are likely to want to index? There are probably only a few of your Tag Types that would need this customization, which (if it works) is a lot less work than custom modifying every sentence or every place. The best way to see if this will work is to try it. Let us know, as I am now curious myself.

Share this post


Link to post
Share on other sites
I have not tried this, but you might test using the location variables rather than text in a modification to the Tag Type sentences. Then you don't have to deal with individual places. This is a guess, but perhaps something like:

[iNDEX:]places:[L6],[L5],[L4][:INDEX]

added to sentence templates that you are likely to want to index? There are probably only a few of your Tag Types that would need this customization, which (if it works) is a lot less work than custom modifying every sentence or every place. The best way to see if this will work is to try it. Let us know, as I am now curious myself.

 

Thanks, Michael. Funny you should suggest that, since I had tested the same thing with the Marriage tag. I had the sentence variables as:

[:CR:]He married [PO] <[PARO]> <[D]> <[L]> <[M1]> <with [WO] present>[iNDEX:]Places:< [L6]:>< [L5]:>< [L4]:>< [L3], >< [L2]>[:INDEX]

and it did seem to work perfectly well except for two problems.

 

The first was that I have a lot of custom roles so there would be quite a lot of work. Not nearly as much work as hand-indexing all 6,000 plus places, though! So that problem is relatively trivial.

 

The second is that if you have BOTH Names and Places indexed automatically, and another manual index as above, then the Names and Places and the third index are output separately. However, if you have only one of Names or Places indexed automatically, and another manual index, the two indices merge together. There may be a way of outputting the indices separately, but I've not found it.

 

I hadn't raised this issue in my original post in order to keep things simple. If I can just get TMG to do what it says on the tin then both these issues disappear :rolleyes:

 

Hope this helps, Michael, and thanks again for your input!

 

Best wishes

 

Tom

Edited by PiercysEtc

Share this post


Link to post
Share on other sites

You are already outside my personal experience, but I noticed a posting here in the Frequently Asked Questions forum under the Reports area. It talks more about the separate and combined indexes and may help explain what you are seeing.

Share this post


Link to post
Share on other sites
You are already outside my personal experience, but I noticed a posting here in the Frequently Asked Questions forum under the Reports area. It talks more about the separate and combined indexes and may help explain what you are seeing.

Thanks, Michael. Sorry about the delay in replying; I got dragged away (don't ask!)

 

Yes, it was because that article said that it was possible to generate a conventionally formatted Places index using the automatic place indexing that I have spent so much time in the last few weeks trying to get it to work. As the article says:

 

Place Related: Places (See the note about places below)

 

Places Index

Places:subfield 1:subfield 2, etc.

Places:Massachusetts:Boston

and
Places

 

You may choose to create multiple index entries for a single place by checking the "By City (L4)", "By County (L5)", etc., options on the Index(es) tab of the Publication Tools dialog. When you check one of those options, you are requesting an index entry for each place that includes that level of detail as well as all larger levels of detail. The subfields may be ordered smallest elements first or largest elements first. In the examples above, I assumed that only the "By City (L4)" option was checked. I also assumed that the "Largest Elements First" option was checked.

In the example quoted above, subfield 1 is Massachusetts and subfield 2 is Boston; the indexing is 'By City (L4)' and 'Largest Element first' is selected.

 

However, when I apply this configuration, I get the equivalent of 'Places:Massachusetts:USA'. In other words, the index levels for me in TMG7.04 appear to go from small to large, not from large to small as the article claims and as you would expect from conventional indexing. There may be occasions when you would wish to find all the incidences of the USA in Massachusets but I am sure there are more occasions when you would wish to find all the incidences of Boston in Massachusets. :wacko:

 

So either TMG DOESN'T do what it says on the tin and the writer got it wrong (unlikely, since it is an 'official' TMG post), or a bug has been introduced into TMG since the article was written four years ago (again, unlikely since it surely must have been picked up by now by the vast TMG user base) or TMG DOES do what it says on the tin (much more likely, surely) and I have overlooked something obvious. All very frustrating

 

Thanks again, Michael

 

Best wishes

 

Tom

Edited by PiercysEtc

Share this post


Link to post
Share on other sites

This is a thank-you, Michael, for the support you have given me in trying to resolve this problem.

 

There have been no further responses to my postings either here or on the Rootweb TMG list-serve so I've escalated the call direct to Wholly Genes support.

 

I'll let you know their solution.

 

Best wishes

Share this post


Link to post
Share on other sites

You are most welcome, Tom. Sounds like you are taking the appropriate next step. I am one user that has not created many indexes so could only do what you are doing and experiment with various settings to see what results occur. The absence of responses sounds like there are few other users with experience in this area either. Good luck, and looking forward to hearing what WG has to say.

Share this post


Link to post
Share on other sites

I see what you are saying...

 

If all of your places have the Country level used and you only select Country level and select Largest element first, then the index would only have countries and higher levels listed subordinate to the country entries. The lower levels don't appear in the index. Higher levels only appear with the lower levels.

 

So if you select city, county state and have Largest element first selected, you get:

 

City

State, County

County

State

State

 

You can't select only State and get:

 

State

County, City

I don't know if this is the design or a bug. It appears to be the design.

 

Not much is described about indexing in Help but it does exactly what is described in help.

 

So, to clarify my query, what is the correct way to order the Places index into the conventional format as below?

..Country

....State, City, etc...

I can't see any way to do that.

 

If I can just get TMG to do what it says on the tin then both these issues disappear...

What are you referring to specifically? I can't find anything that WG says the indexing will do that it doesn't do. All descriptions of the indexing are minimal at best.

Share this post


Link to post
Share on other sites

Thanks for your reply, Jim.

 

I see what you are saying...

 

If all of your places have the Country level used and you only select Country level and select Largest element first, then the index would only have countries and higher levels listed subordinate to the country entries. The lower levels don't appear in the index. Higher levels only appear with the lower levels.

 

So if you select city, county state and have Largest element first selected, you get:

 

City

State, County

County

State

State

 

You can't select only State and get:

 

State

County, City

Yes, this is what I have found.

 

I don't know if this is the design or a bug. It appears to be the design.

 

Not much is described about indexing in Help but it does exactly what is described in help.

Maybe so, Jim. The Help files may be unclear, but:

 

1 The link in the FAQs that Michael referred to says:

 

Places Index

Places:subfield 1:subfield 2, etc.

Places:Massachusetts:Boston

and later the clarification

 

I assumed that only the "By City (L4)" option was checked. I also assumed that the "Largest Elements First" option was checked.

2 The TMG Users Guide, UK Edition, says on p283:

 

Places

Places will index By short place, By place detail, By village,

By town, By county, or By country.

 

Check Largest Element First to place the largest place

element in the first position. For example, if this option is

checked, an index entry would appear as

 

Cambridgeshire

Wyton, Huntingdon

instead of:

 

Wyton

Huntingdon, Cambridgeshire

So it seems that Wholly Genes intend at least one configuration to index in conventional format.

 

It's simply that I have not managed to find that configuration.

 

So, to clarify my query, what is the correct way to order the Places index into the conventional format as below?

..Country

....State, City, etc...

I can't see any way to do that.

I can't either. I'm sure that the TMG support/help desk will be able to, though. I've not heard from them yet, but I'm sure they will sort out where my configuration is going wrong.

 

If I can just get TMG to do what it says on the tin then both these issues disappear...

What are you referring to specifically? I can't find anything that WG says the indexing will do that it doesn't do. All descriptions of the indexing are minimal at best.

The descriptions in Help, yes, I agree. However, the description in the two items I have quoted above suggest that the conventional format for a places index CAN be achieved with some configuration of the automatic indexing available in TMG - if you don't mind my referring to them as 'what it says on the tin'.

 

I also find it nearly impossible to believe that TMG, which has such a reputation for doing things right, would not implement a conventional places index among the options. It's almost like driving your new car back from the forecourt and finding that the steering was 'unconventional' - the car turned left when you turned the wheel right... That doesn't sound at all like a TMG feature!

 

Best wishes

Edited by PiercysEtc

Share this post


Link to post
Share on other sites
So it seems that Wholly Genes intend at least one configuration to index in conventional format.

 

It's simply that I have not managed to find that configuration.

That's because it won't do it.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×