Jump to content


  • Content Count

  • Joined

  • Last visited

Community Reputation

0 Neutral

About milefaker

  • Rank
  1. I have a table Subjects containing a list of unique values in field SubjectName. This field serves as a the value list for populating a field SubjN in a second table Articles, which is related to Subjects. My problem is that when I make changes to the values in SubjectName, the corresponding contents in SubjN do not refresh. Is there an automatic way to refresh them? There is a different field in Articles that populates automatically depending on the key that relates Articles to Subjects, so I have used only a value list for SubjN. Thanks for your suggestions.
  2. I have a table Categories, containing fields CategoryNameCategorySerial#CategoryDescription. There is a second table Subcategories, containing fields SubcatNameSubcatSerial#CatSerial#Data The tables are linked by way of the fields CategorySerial# and the key field CatSerial#. Problem I would like the serial numbers in SubcatSerial# to "restart" from 1 within each Category, rather than simply having a unique SubcatSerial# for each record in Subcategories. Using automatic serial numbers in SubcatSerial# is clearly not the way to do this, but am not sure what will work. Thanks for your suggestions.
  3. The calculation is _coll & "; " & _synt & "; " & _Objects & "; " & _M in which the four items beginning with an underscore are names of fields in the current layout. As mentioned, this string-concatenation works when I use Replace Field Contents manually. I have indeed been using Go to Layout as the step before Replace Field Contents in the script, as Ben Goldstein suggested. I've just found that I can avoid the error by including the name of the layout before each field-name: Currentlayout:_coll & "; " & Currentlayout::_synt & "; " & Currentlayout::_Objects & "; " & Currentlayout::_M So the problem is solved. But it's curious to me that so much more specificity is needed in the script than in manual operation of the same command. Thanks for your help.
  4. Actually, I'm getting the error message while writing the script; I'm unable to save it, much less run it. The "Replace Field Contents" dialog that opens when I click the lower of the two "specify" buttons will not allow me to save the calculation I am using, even though that same calculation works manually.
  5. Using "Replace Field Contents" to replace with a calculated result works fine manually. But when I place it in a script and try to specify the same calculation, it always gives me the error "The specified table cannot be found", regardless of what table I have selected. What am I doing wrong? Thanks!
  6. Attached is a zipped clone. I've removed as much extraneous matter as possible, while leaving all the current category fields intact for your reference. Thanks for your patient help. I've described the goals and structure in a previous post in this thread. A few other points: I ordinarily work while browsing the db as a table, which allows me to see the greatest number of records at once. However, because there are so many fields, I keep those I'm not working with collapsed so that they don't take up so much space. Fields that represent semantic categories are ordinarily named beginning with a capital letter; those that have other functions ordinarily begin with an underscore. Clone.fp7.zip
  7. Yes, and unless I switch to a different structure there will be more coming. This database contains what will eventually be a thesaurus. The fields in question correspond to the larger semantic categories of the thesaurus. Those categories are parallel to each other; that is, there is no meaningful hierarchical relationship among them. However, the records within each category fall first of all into subcategories and secondarily within the subcategories into a particular order of presentation. I indicate those subcategories and the presentation-order as the sortable contents of the category-fields. That is, each category-field contains a distinctive set of sorting orders for the relevant records. A few other relevant details: The relationship of records to categories is one to many. Any given semantic category typically contains relatively few records (rarely more than 100 at the moment and in the future these numbers will drop, as categories become more and more finely divided). Any given record may need to be sorted by more than one sort order, depending on which of the categories it belongs to is being treated. So I have not found a way to assign an absolute "sort-ordering" code to each record, which might have simplified everything. The semantic structure of the eventual thesaurus is not yet settled. I am constantly revising the sort orders of different semantic categories, and moving records in or out of different semantic categories, and dividing or combining semantic categories. I have one field that concatenates the contents of all the semantic category fields, so that I can search quickly for a given category, without having to hunt manually through a forest of fields. The system I am using, though strange, gives me great flexibility in the tasks I've just described. I realize that hundreds of fields in one layout is unconventional and messy. At first, I tried assigning each category to its own layout, but with hundreds of categories it was unwieldy. Doing it this way has proven the least trouble so far, and far more effective than trying to do the whole thing in a text editor for a number of reasons. But I'd be very happy to hear suggestions for other approaches.
  8. I see what you mean. However, a single headword is not a single word, merely a single form. To continue with the example of "can", once I create a record in the Definitions table for "can" in the sense "able" and another record for "can" in the sense "metal container", I want them to remain separate, and I want to see only the history of each separate "thread" of development. Otherwise, things could get extremely confusing.
  9. Thanks. The second suggestion worked beautifully. (With over 300 fields currently, I was loath to do with manually.)
  10. Is there a way to select a subset of the field names in layout view and have them alphabetized automatically? Or must that be done manually? Thanks.
  11. Thanks for your quick reply. No. Headword to definition is one to many. I mean that "definitions" contains a field whose contents are populated from "headwords". That way, there is no danger that I will accidentally alter that content manually when working in definitions, and if I need associated material from headwords, it can also be used to populate a field in definitions. How would I know that any two records represented different versions of the same material? Thanks for the reference.
  12. Can anyone suggest a way to implement row journaling? I.e., keeping past versions of a given record accessible but never displaying them, unless for some reason I want to go back and examine long-discarded changes. I can already do that by consulting backed up versions of the whole database, but it's hard to know which backed-up version to look in for something I want to find. This database is being used to compile dictionary data. There are two main tables: "headwords": contains individual words that have unique spelling, together with information about their frequency of occurrence, sources that contain them, and graphic variants."definitions": contains definitions, parts of speech, notes on usage, and so forth. Records in "headwords" are related one-to-many to records in "definitions", because a single distinctive word form may have unrelated meanings and parts of speech. So for example, "can" appears only once in "headwords", but several times in "definitions", each time with all the things that make the various words spelled "can" different from each other. I use the serial number from "headwords" to look up word forms and populate the appropriate field in "definitions". Content is added to "headwords" occasionally, but there is very rapid and repeated alteration of the material in "definitions", and I would like to be able to review past changes of that material at will. In "headwords", each headword field has unique contents. But in "definitions", no single field has unique contents except the serial number. Thanks for any suggestions or references.
  13. I see. That clarifies a number of things I haven't understood. Thank you for your persistence — it's now time for me to get down to actually using my database for the next 8 hours — but I'll be back on this problem again tomorrow.
  14. Ah, that's very interesting! I see the BOM as the first character of the file. Works like a dream. I'm glad to hear that even without a BOM, FMPv10 will do this correctly. For now, however, I've set my BBEdit preferences to save everything as UTF-8 with BOM. Am now deleting the files I had placed on-line. Thanks for your labor!
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.