George Papadakis Posted April 28, 2005 Share Posted April 28, 2005 Hi, I have this problem: I converted my databases from FM6 to FM7. (Mac OS X). Thousands of records are wtitten in Greek language and in the previous, non Unicode text. The conversion is OK but now I have find problems: Due to the fact that other records were written with other greek keybords (there are 3 different greek keybords: 1Greek, 2Greek old, 3Greek separated) when I search something, FM7 does not find every record of that. If I type with different greek keyboard I get different results. With other words I can't find everythng that exists in my database If I retype something then FM7 finds it at once. It seems that I have to retype every text in every field in every record. This, of course, is impossible (or not so clever I think). Is there any idea? How can I make a text conversion like this? Thanks Link to comment Share on other sites More sharing options...
FileMakin' Tom Posted April 28, 2005 Share Posted April 28, 2005 I dunno. It's all Greek to me. (Sorry - just could not help myself with such an obvious invitation for a pun). Good luck and perhaps another of our Greek contributors can assist you. Be well . . . Tom Link to comment Share on other sites More sharing options...
comment Posted April 28, 2005 Share Posted April 28, 2005 You need to tell FileMaker to "re-type" the texts. It's still a lot of work, but... First create a global text field gField. Next, define three calculation fields (result is text) = Substitute ( GetField ( gField ) ; [ "oldAlpha" ; UnicodeAlpha" ] ; [ "oldBeta" ; UnicodeBeta" ] ; .. [ "oldOmega" ; UnicodeOmega" ] ) You may have to copy/paste the non-Unicode characters into the formula. You need three such calculation fields, one for each old keyboard. Now find the records that were entered with the 1Greek keyboard. Enter the name of the first field into gField and replace the first field's contents with the result of the first calculation field. And so on - hopefully the pattern is clear. Using scripts to automate parts of the process, should take about an hour. Be sure to have a backup, before using Replace. Link to comment Share on other sites More sharing options...
FileMakin' Tom Posted April 29, 2005 Share Posted April 29, 2005 How cool, Comment. (Round of applause). It is from those like you that the extension of FMP techniques are expanded and shared. My thanx from the FMP community. Tom Link to comment Share on other sites More sharing options...
George Papadakis Posted April 29, 2005 Author Share Posted April 29, 2005 Thanks, Thanks, Thanks, I didn't test yet but I feel this is the solution. Thanks again! Link to comment Share on other sites More sharing options...
comment Posted April 29, 2005 Share Posted April 29, 2005 Let us know how it goes. Now that I look at it, it can be streamlined a bit: you can combine the three calc fields into a single one = Substitute ( GetField ( gField ) ; [ "oldAlpha1" ; UnicodeAlpha" ] ; [ "oldBeta1" ; UnicodeBeta" ] ; ... [ "oldOmega1" ; UnicodeOmega" ] ; [ "oldAlpha2" ; UnicodeAlpha" ] ; [ "oldBeta2" ; UnicodeBeta" ] ; ... [ "oldOmega2" ; UnicodeOmega" ] ; [ "oldAlpha3" ; UnicodeAlpha" ] ; [ "oldBeta3" ; UnicodeBeta" ] ; ... [ "oldOmega3" ; UnicodeOmega" ] ) If you are bold (and have a backup) you can try putting all the text fields on a layout, then script a loop - see attached demo. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.