GenoPro Support Forum

So, close but not quite close enough yet!. There are some new errors here but also some elementary slips on my part.
I have posted another attempt at the 2.0.0.6 skin in the first post of this thread. (RC6)
Update: This version is now availlable in GenoPro 2.0.0.6 download

I have also included as a bonus the optional translation of Occupation Titles. But this bonus comes at a price. I have changed the format of NameDictionary so that all attributes relating to Places should now be prefixed with 'P.' and attributes relating to Occupation Titles must be prefixed with 'O.' The NameDictionary.xml file included with the skin has examples of each. This is to avoid confusion when surnames are the same as places and occupations, e.g. London, England, Baker, Plumber etc. Whilst you may want to translate places and occupations you do not necessarily want to translate surnames as well.

Similar to Place translation, the Config.xml Parameter LangOccupation specifies the language code for Occupation Titles in the .gno file when they are to be translated to the skin language.

This is a Great idea.  That space could be used to display additional information, tips or some extra help regarding the control having the focus.  I am wrapping up the multi-lingual version, and later plan to add "tips" for each control.  This way, the user won't have to read the user's manual to discover hidden features of GenoPro.

Here's a few more suggestions. As you said, it's coming to an end as I'm quite happy with the result. I think now you may/will have more requests from other language users. Also waiting for the trans implementation of occu, ...

A) Translations

1) The Fem gender doesn't work (was OK before):
 

Note: it's OK when <fHideFamilyDetails = "N"/>

B) Bugs, corrections & additions

1) with the param <fHideFamilyDetails="N"> in config.xml, the said param cannot be changed within the config setting panel prior to generate a report.
I got this for one of my Ind (after selecting yes in the panel):

Clicking on the red ul "Unknown", I got this:

Strange!
For info, I didn't try changing the other params. I don't know if they're working or not?

2) Generated Place.htm
- I think a space should be removed in-between ancêtres and à:
<meta name="description" content="Historique de famille et ancêtres&nbsp; à Glasgow" />
- a lookup substitution should also be implemented:
<meta name="keywords" content="Lieu United Kingdom" />

3) <FmtMetaDescIndividual1>
A NameDic look up should be implemented, as shown in one of my generated Ind.htm (b.Place):
<meta name="description" content="Rapport ancestral de Martine Bronca, fille de ..., née le 18 octobre 1948 à United Kingdom. Martine a un époux/partenaire nommé ..." />

4) <PhUnion>
A check-up routine, similar to GodFathers and Godmothers, should be implemented, for multiple witnesses.
I added a param {9} in Lang.vbs and it's working. This extra param {9} will simply be ignored in an EN report:

' The family has at least one union/marriage
 g_RegEx.Pattern = "[,&]"
 For Each m In collUnions
  Report.WritePhraseDic "PhUnion", ..., strAlso, g_RegEx.Test(m.Witnesses)
  ...

For info, in FR, "witnessed by ..." is "le témoin était ..." or "les témoins étaient ..."

Note: I also suggest Genopro to implement a message system when the cursor is in some field with an input particularity. In our case, the Witnesses field message could say, e.g. "Use ',' or '&' for delimiters in-between names". This message could pop up in the following red rectangle area:

or in a bubble.

5) 1st letter capitalization of the Phrases appearing in Timelines bubbles?
It's OK now for the 1st Ph but not yet for the subsequent Ph, e.g.She was born ... . she died ...

6) my post 17791 (error message) has been implemented but only for the ERs. For the SRs, the message still pops up.

7) my post 17793 (adopted/fostered child) has been partly implemented. That's OK. In addition, a space before "through" should also be added in <PhPL_foster>.

8) Google message

For info, the place substitution is not done on United Kingdom when input as a Country.

Also on clicking on "Afficher la carte Google ...", the message is not 100% Unicode compatible. The red ul should be '.

Thank you Ron.
JC

Edited: Monday, June 25, 2007 by jcguasp

A.

1. 'Custom' in FmtHomeSkin. an extra space is needed when 'custom' is added. I have left 'as is' as a 'formatted string' rather than a 'phrase' because of the problems adding html codes to Dictionary 'phrases'. A 'formatted string' does not have conditional subphrases and so the extra space cannot be handled in the phrase.

Your '2007.06.18' is the 'SkinVersion' taken from Config.xml. It just so happens it is expressed as a date, but need not be. I tend to prefer dates in the format yyyy.mm.dd to avoid confusion with the american format mm.dd.yyyy and the British (& European?) format of dd.mm.yyyy in dates such as 12/04/2007 which could be 04/12/2007 depending on which side the 'pond' you live. If you want to change the Skin version format I suggest you amend config.xml. FmtDateSort is purely for formatting dates prior to sorting and so is not appropriate here.

B.

1. Thanks, I have corrected it now. Note that the Place description is no longer part of the phrase as it is now handled in the code.
2. Again code amended.

My interpretation of an acqaintance is someone you know reasonably well, i.e. you have a proper conversation with them. The Internet Dictionaries suggest someone less intimate than a friend. Whereas 'knowing' someone is a weaker relationship, perhaps you know their name and only exchange an ocassional greeting. There is a English expression 'nodding acquaintance' meaning you just acknowledge each other say in the street.

3. Code fixed

4. I had overlooked this one, so thanks for the prompt, code should be OK now.

5. I have amended the phrase to add the space before 'through'. As regards reorganisation of the rest of the phrase, this requires a little more effort so I'll leave it untill 2.0.0.7. Also I believe there is a problem with GenoPro's interpretation of half siblings. My view is that half siblings are biologically related via a single parent, and so adopted/fostered siblings should not be included.

6. Code amended to take account of gender of mate(s).

7. All sorted. I think strTitle in source.htm was redundant so I have deleted it.

8. All fixed.

9. I have fixed using Util.StrFirstCharUCase

10. I prefer to  extend 'Custom markup' to handle these translations rather than using ther NameDictionary. These are mostly unique strings appearing once in the .gno file. I am considering allowing for conditional text formats e.g. [{?en}engineer][{?fr}ingénieur] in these fields. The Narrative skin would then select the required version.

And finally, you can prevent Custom Tags from appearing in the Additional Items section of a report by using the Tag Editor to place them in a Custom Tag Dialog Layout with a description starting with an underscore (_) or give the Tag itself a description starting with underscore. The Dictionary-defined 'Private' string can be used in place of an underscore.

Also remember that you can, and should,  change the default Custom Tag Layout Description, ' A custom tag is a generic placeholder....' to something more suitable.

A few more issues related to translation or to the rep itself.

A) Translations

1- Report.WriteFormatted Dic("FmtHtmlHomeSkin") in home.htm
- It would be better if & " ", after Dic("Custom"), could be removed. The space can easily be provided within the template.
- Is there a way to change/customize the version date at the very end of the ph, i.e. D, M & Y:
Ce rapport généalogique a été généré par GenoPro® version 2.0.0.5 du 20.6.2007 avec la matrice de traduction personnalisée Rapport narratif français version 2007.06.18.
For info, I manually changed the params order for the middle date (20.6.2007). However it would be good, generally speaking, if all dates could automatically match the template <FmtDateSort> in dic.

B) Bugs, adjustment or additions

1- <PhPlaceDescription> in dic
The Ph doesn't seem to be working as there's no narrative in the report.

2- Sub WriteHtmlRelationships(i) in util.vbs
The enum (2 ways SR) shows: "Roommate", "LivesWith", "Neighbor", "Acquaintance"
I think the following should also be added:
"Associate", "Know", "Relative" and "WorkWith"
By the way, as already mentioned in a previous post, what is the difference in-between Acquaintance and Know?

3- Place Index alpha sorting
- Despite the place name substitution via NameDic, my place "Royaume-Uni" (United Kingdom in EN) is not alpha sorted in the Place Index:

4- my post 17791 dated 20/5 (error message) has not been implemented yet.

5- my post 17793 dated 20/5 (adopted/fostered child Ph order and missing siblings) has not been implemented yet.

6- <<FmtMetaDescIndividual2>
It seems that param {2} always default to <_Spouse>, ignoring ends _M or _F.
I cannot get any Fem word in the meta Desc Nar Ph.
Please check the routine.

7- Few bits & pieces not written to generated .htm files. I May be wrong or the var may be = "" though.
- FmtCopyright in default.htm
- <meta name="description" content="??" in place.htm
- "Content-Language" content="fr" in popup.htm
- strTitle from @[ Report.WriteText strTitle ]@ in source.htm
- svghelp.htm: shouln't everything (<Title> and text) be in Dic?

8- Minor cosmetic corrections (incl. some words from comments)
- from PLaces.SortBy("Name") to Places.SortBy("Name") in places.htm
- from preposiiton to preposition in util.vbs line 40
- from shown on on to shown on in config line 37
- from identivier to identifier in dic line 478
- <PossessiveProperNoun> (for update purpose)
  from T="(^[aeiouyh].*$)=d'$1'.*$)=de $1:"
  to   T="(^[aeiouyh].*$)=d'$1.+$)=de $1:" in dic line 625 (the 1st version didn't work, even with the 2nd ' removed)
- from pupose to purpose in namedic line 3
- from atribute to attribute in namedic line 12

9- Is it possible to capitalize the 1st letter of the Ph appearing in Timelines bubbles? I tried to add up {\U} in the templates but it shows 4 square chars!

10- It would be useful to have the NameDic lookup substitution extended beyond the places, i.e. for occupations, cause of death (specify), picture name, source subtitle, brief source description, citation ref., text quoted from source and the many comments. Plenty fields then!

Finally, could you remind me/us how to avoid the appearance of this in a rep:

Thank you Ron and good work,
JC


Edited: Thursday, June 21, 2007 by jcguasp

I have now posted the fourth update to my 2.0.0.6 release candidate. As usual this can be found in the first post of this thread.

Hopefully I have corrected all the points raised since 2.0.0.6RC3, but if I have overlooked any please don't hestitate to prompt me via this thread or a PM via this forum.

Update: I forgot to mention that the changelog.txt file for 2.0.0.6RC4 has two sections. Section 1 details changes between 2.0.0.5 and 2.0.0.6RC3 and Section 2 documents changes between 2.0.0.6RC3 and 2.0.0.6RC4. Therefore if you have already made a translation based on RC3 you only need to view Section 2.

One adjustment to do when a NameDic lookup is involved. Here it's about a place:
The entry in my nameDic is <N EN="United Kingdom" FR="Royaume-Uni" FR_L="&#32;au [Royaume-Uni]" />
The input in my .gno file is "United Kingdom".

1) Royaume Uni should be listed under the R letter in these indexes:

2) UK should be substituted by Royaume Uni within the Most popular places list:

Hope I didn't forget any other loc, within the rep, for these substitutions. Thank you Ron,
JC

2) and 3) I didn't know of their existence, I should be all set with these...

rgds,

Ronald

Some responses to the recent postings:

Firstly JC,

A 1) PlacePrefix I think this is answered in my earlier post. the address components are not GenoPro places so have no category or preposition. The solution I intend to use is to reword the phrase as indicated.

a 2) Metadata. Agreed fix required.

B 1) NamePossessive. Not sure what is happening here. Could you try changing <PossessiveProperNoun T="(^[aeiouyh].*$)=d'$1.*$)=de $1:" /> to <PossessiveProperNoun T="(^[aeiouyh].*$)=d'$1.+$)=de $1:" />?

b 2) FmtDatesIndividual. Yes you are correct. I'll fix it, thanks

and Ronald,

1) PossessiveProperNoun. Does the Dictionary entry <PossessiveProperNoun T="(^.*$)=de $1:" /> solve your problem, as suggested by Gerardo and also used as a French example in the Dictionary? Or do you sometimes need the Dutch possessive form, 'Rons',  and at other times 'de Ron'?

2) The 'custom markup' feature already allows a <?merge "some text ............" customtagname ?> format that can be used in the Place.Comment field. Also if you create custom tag called _Narrative this will be added after the Place description and can contain 'Custom Markup' tags.
I will also change the code so that Place.Description can also contain 'Custom Markup'.
See http://support.genopro.com/FindPost17355.aspx for more on Custom Markup. I did recently find a bug in the <?merge?> handling so you may want to wait for my next release before trying it out.

3) Custom Tag Layout Descriptions. The sentence you quote is the default Layout Description which can of course be changed using the Tag Editor. For this reason there is no need for a Dictionary translation. Indeed since the user can edit the text to anything he/she requires and also have more than one layout then Dictionary translation is impossible. I can see no reason why anyone would want to leave this default description in their report as it is completely meaningless in the context of a report. I would prefer that this default phrase was removed from GenoPro and the initial description left blank or perhaps set to "<Enter a description here>".
I am considering adding another feature to Custom Markup so that text fields can contain more than one language version of the text and the version selected for the report will be that corresponding to the  Skin language.
The two levels. i.e. layouts and custom tag descriptions, allow custom tags to be categorised and grouped separately, using the layout description to describe the group, a very useful facility.

JC & Gerardo,

I'll work through these suggestions.

and lastly, but not least, to maru-san

PhGodparents. Since Godfather & Godmother are simple text fields I will alter the code to detect either a comma or ampersand in the text and then assume multiple godparents, passing a couple of boolean parameters to indicate more than one godfather or more than one godmother, and change 'is/was' accordingly.