great, thought I saw it all in one, but that statement is trivial it can go up in there with my cwcem = cwcem or 0 statements.

rmn = string.sub(fhGetTag(ptrWork), 1, 1) table.insert(tblRMN, rmn) This is a slice of my matCEM function...It does a lot of stuff. Many times, one may only need a table or two out of its many tables. so, I know I have seen tatewise do this when perusing his code and never understood it, and I cann...

Well here is the thing in its penultimate glory the facts function is now rtvPtrFromINDI() although I have not guarded any other thng than an INDI or a linked SOUR, I have guarded against a naked SOUR I did it by taking the cowards way out, setting everything to null and chugging merrily along. I sa...

embedded is a code snippet I am working on, and have been for two weeks spare time now. although it is not clear here at this point in the code, I will take in either an INDI ptr, or a SOUR ptr. trivial to figure out which, can do that with fhGetTAG. INDI ptr needs to find out if a SOUR exists in th...

I had not thought there would be a difference between nil and an empty string.

But yeah, I have to do a great deal of flipping back and forth between " ", 1947, 1947 so a function for that seems the way to go.

Thanks.

BLANKS = rep(' ', 4)

BLANKS is correct

reading the documentation near as I can figure it out fBirth bein nil, Birth should be BLANKS but it is not.

under what circumstances does the or trick work? fBirth will either contain a number or ''.

what I have learned this weekend after another all day and night weekend of debugging, normalizing code, putting it in functions (since I cannot find a /copy sort of text(code) into the program *here* thing) and not ready to figure load and . : yet, I now have a great deal of functions that change c...

In fact that debug will never work (well in once case if 1 is in the table ): there is no way I see to get the return value from a pcall, or xpcall without a great deal of code complication and dependency. that is, rcdid = sts, err( pcall( fhgetrcdid, ptrindi)) rcdid, sts, err pcall(fhgetrcdid, ptri...

I have went thru and inspected all my code and did minor fixes to standardize how birth and death are calculated I now chose mid.birth and mid.death mid.birth seems to be pretty accurate. even way out mid.death is often BLANK (transformed) lifedates2 is wild for the scant info profiles. so, at prese...

Ja, thats quite a bit smaller and simpler. and I checked it, put it in the stupidest possible place in my program, right after the -- [[ ]] comments before anything was done and it QUIETLY did not work, not interrupting the debug session, so I know the portable debug is not right, either I did not c...

So, I have been doing a lot of debugging lately, and tire of typing in stuff all the time because generally (except possibly for Jane and Tatewise) we do a lot of stuff with INDI records. tatewise made a quick debug statement or two for me in the process of explaining something, and I thought, yanno...

So,

function
end

local fred

function
end

is the same as putting fred at say the top of the file?

the reference manuals for LUA are a work of horror. They may occasionally get a very simple idea across in 20 pages, but nothing advanced in there, nearly completely unintelligible.

I was so intent on capturing the clc as (x00) that I didnt think about the fact I rebuilt it, simplifying the thing enormously, which caused everything else to go upside down.

it appears that lifedates2 isnt much of a program

I think the best solution will be to use mid.birth as birth and late.birth as death.

because by this time, I only have 2 digit multipliers. if I have missed some case, it should stick out like a sore thumb yeah, hints are not instructive, its very plain I dont understand. tell me what is wrong with this "^(.-) (%(x(%d%d)%) .- (great%-) (.*)" not learning a thing that shoul...

Mike Tate: I don't think it is possible to match both of your examples with a single pattern. me: ok down to two general cases. which is once case in general half great (x10) great-uncle (can be anything) aunt, nephew, uncle's wife's cat's girlfriend great (x10) great-uncle (can be anything) aunt, n...

this requires expertise: -- great- add 1 to (xNN) aunt uncle etc.. pre, clc, one, pst = string.match(relation, "(.-) ?(%(x(%d+)%) ('great-') (.-)") if one ~= nil then if clc ~= nil then gen = tonumber(clc:gsub("(%d+)")) + 1 clc = "(x" .. gen .. ")" else clc = ...

I am not getting rid of removed I am consolidating the two versions it comes either 1st cousin once removed or 1st cousin once-removed I am removing the minus in the specific case it only happens on that exact type near as I can tell, I generalized it because I did not want to exhaustively understan...

Solved in a weird way, now I know the difference between (.-) and (.*) and it worked to shut off numbers on a greedy capture as well. -- correct relation sort relation = relation:gsub( "-removed", " removed" ) -- first relation = relation:gsub( "^(%d[^%d])(.*)", " ...

-- correct relation sort relation = relation:gsub( "-removed", " removed" ) -- first ^^ FH not consistent sometimes (once removed) sometimes (once-removed( relation = relation:gsub( "^(%d[^%d])(.*)", "0%1%2" ) -- second fixup front of string only to2 digits re...

I misunderstood the anchor, thought I was anchoring to the x.

that explains it

dont remember e birth but lbirth was 1849

relation = relation:gsub( "^x(%d)", "x0%1" ) is what will be my final production statement when I get this handled. I havent misunderstood anything. I am debugging as those dont appear to work I set a breakpoint after each statement executed to inspect the contents of relation re...

you would think so, now wouldn't you? -- correct relation sort relation = relation:gsub( "^x(%d)", "x0%1" ) * relation = relation:gsub( "^x([2-9])", "x0%1" ) * relation = relation:gsub( "^x(%d[^%d])", "x0%1" ) * debug = debug the stars repr...